Compare commits
26 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 5273cf6f03 | |||
| 3f7a5dd5a1 | |||
| 146efdf6bd | |||
| db5b6bb6c7 | |||
| 35df07725e | |||
| 867a0a3eaf | |||
| a647179e72 | |||
| 438fba347a | |||
| 8066fc4386 | |||
| 9cd9ccd8e0 | |||
| a9cbd1b2ba | |||
| 685582b7e6 | |||
| 961322f8ba | |||
| b40e890e2b | |||
| 9525d56ffc | |||
| 5433605fec | |||
| b981e30f46 | |||
| e2d35b6fa0 | |||
| 45190980a9 | |||
| 627148a87f | |||
| 74be7688c9 | |||
| 7a2b1da9ff | |||
| dfd7234fd6 | |||
| 22b1a78132 | |||
| d28ad1a292 | |||
| 1b22718d5a |
1006
data/categories.json
1006
data/categories.json
File diff suppressed because it is too large
Load Diff
14
data/config.json
Normal file
14
data/config.json
Normal file
@@ -0,0 +1,14 @@
|
||||
{
|
||||
"site_name": "ParamHub",
|
||||
"site_subtitle": "参数百科",
|
||||
"footer_text": "ParamHub - 模型与硬件参数速查平台",
|
||||
"icp_number": "",
|
||||
"copyright_year": "2026",
|
||||
"contact_email": "wlq@tphai.com",
|
||||
"github_url": "",
|
||||
"llm_base_url": "https://open.bigmodel.cn/api/paas/v4",
|
||||
"llm_api_key": "2259e33a1357460abe17919aaf81e73d.K44a8LPQTmFM5PKm",
|
||||
"llm_model": "glm-4.5-air",
|
||||
"llm_vision_model": "glm-4.6v",
|
||||
"updated_at": "2026-04-27 23:58:26"
|
||||
}
|
||||
156
data/cpus.json
156
data/cpus.json
@@ -1,10 +1,150 @@
|
||||
[
|
||||
{"id": "epyc9654", "name": "AMD EPYC 9654", "manufacturer": "AMD", "architecture": "Zen 4", "cores": 96, "threads": 192, "base_clock_ghz": 2.4, "boost_clock_ghz": 3.7, "l3_cache_mb": 384, "tdp_watts": 360, "price_usd": 11000, "release_year": 2022, "description": "AMD顶级服务器CPU，96核心"},
|
||||
{"id": "epyc9554", "name": "AMD EPYC 9554", "manufacturer": "AMD", "architecture": "Zen 4", "cores": 64, "threads": 128, "base_clock_ghz": 3.1, "boost_clock_ghz": 3.8, "l3_cache_mb": 256, "tdp_watts": 360, "price_usd": 6800, "release_year": 2022, "description": "64核心高性能服务器CPU"},
|
||||
{"id": "epyc9454", "name": "AMD EPYC 9454", "manufacturer": "AMD", "architecture": "Zen 4", "cores": 48, "threads": 96, "base_clock_ghz": 2.75, "boost_clock_ghz": 3.8, "l3_cache_mb": 192, "tdp_watts": 290, "price_usd": 4100, "release_year": 2022, "description": "48核心服务器CPU"},
|
||||
{"id": "xeonw9359x", "name": "Intel Xeon w9-3595X", "manufacturer": "Intel", "architecture": "Sapphire Rapids", "cores": 56, "threads": 112, "base_clock_ghz": 1.9, "boost_clock_ghz": 4.8, "l3_cache_mb": 105, "tdp_watts": 350, "price_usd": 6200, "release_year": 2023, "description": "Intel顶级工作站CPU"},
|
||||
{"id": "xeonw5345", "name": "Intel Xeon w5-3435", "manufacturer": "Intel", "architecture": "Sapphire Rapids", "cores": 24, "threads": 48, "base_clock_ghz": 3.1, "boost_clock_ghz": 4.7, "l3_cache_mb": 45, "tdp_watts": 230, "price_usd": 950, "release_year": 2023, "description": "中端工作站CPU"},
|
||||
{"id": "ryzen97950x", "name": "AMD Ryzen 9 7950X", "manufacturer": "AMD", "architecture": "Zen 4", "cores": 16, "threads": 32, "base_clock_ghz": 4.5, "boost_clock_ghz": 5.7, "l3_cache_mb": 64, "tdp_watts": 170, "price_usd": 550, "release_year": 2022, "description": "顶级消费级CPU，适合AI开发"},
|
||||
{"id": "ryzen97950x3d", "name": "AMD Ryzen 9 7950X3D", "manufacturer": "AMD", "architecture": "Zen 4", "cores": 16, "threads": 32, "base_clock_ghz": 4.2, "boost_clock_ghz": 5.7, "l3_cache_mb": 144, "tdp_watts": 120, "price_usd": 700, "release_year": 2023, "description": "带3D V-Cache，游戏性能更强"},
|
||||
{"id": "intel14900k", "name": "Intel Core i9-14900K", "manufacturer": "Intel", "architecture": "Raptor Lake Refresh", "cores": 24, "threads": 32, "base_clock_ghz": 3.2, "boost_clock_ghz": 6.0, "l3_cache_mb": 36, "tdp_watts": 125, "price_usd": 580, "release_year": 2023, "description": "Intel顶级消费级CPU"}
|
||||
{
|
||||
"id": "epyc9654",
|
||||
"name": "AMD EPYC 9654",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 4",
|
||||
"cores": 96,
|
||||
"threads": 192,
|
||||
"base_clock_ghz": 2.4,
|
||||
"boost_clock_ghz": 3.7,
|
||||
"l3_cache_mb": 384,
|
||||
"tdp_watts": 360,
|
||||
"price_usd": 11000,
|
||||
"release_year": 2022,
|
||||
"description": "AMD顶级服务器CPU，96核心",
|
||||
"subcategory_id": "server"
|
||||
},
|
||||
{
|
||||
"id": "epyc9554",
|
||||
"name": "AMD EPYC 9554",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 4",
|
||||
"cores": 64,
|
||||
"threads": 128,
|
||||
"base_clock_ghz": 3.1,
|
||||
"boost_clock_ghz": 3.8,
|
||||
"l3_cache_mb": 256,
|
||||
"tdp_watts": 360,
|
||||
"price_usd": 6800,
|
||||
"release_year": 2022,
|
||||
"description": "64核心高性能服务器CPU",
|
||||
"subcategory_id": "server"
|
||||
},
|
||||
{
|
||||
"id": "epyc9454",
|
||||
"name": "AMD EPYC 9454",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 4",
|
||||
"cores": 48,
|
||||
"threads": 96,
|
||||
"base_clock_ghz": 2.75,
|
||||
"boost_clock_ghz": 3.8,
|
||||
"l3_cache_mb": 192,
|
||||
"tdp_watts": 290,
|
||||
"price_usd": 4100,
|
||||
"release_year": 2022,
|
||||
"description": "48核心服务器CPU",
|
||||
"subcategory_id": "server"
|
||||
},
|
||||
{
|
||||
"id": "xeonw9359x",
|
||||
"name": "Intel Xeon w9-3595X",
|
||||
"manufacturer": "Intel",
|
||||
"architecture": "Sapphire Rapids",
|
||||
"cores": 56,
|
||||
"threads": 112,
|
||||
"base_clock_ghz": 1.9,
|
||||
"boost_clock_ghz": 4.8,
|
||||
"l3_cache_mb": 105,
|
||||
"tdp_watts": 350,
|
||||
"price_usd": 6200,
|
||||
"release_year": 2023,
|
||||
"description": "Intel顶级工作站CPU",
|
||||
"subcategory_id": "server"
|
||||
},
|
||||
{
|
||||
"id": "xeonw5345",
|
||||
"name": "Intel Xeon w5-3435",
|
||||
"manufacturer": "Intel",
|
||||
"architecture": "Sapphire Rapids",
|
||||
"cores": 24,
|
||||
"threads": 48,
|
||||
"base_clock_ghz": 3.1,
|
||||
"boost_clock_ghz": 4.7,
|
||||
"l3_cache_mb": 45,
|
||||
"tdp_watts": 230,
|
||||
"price_usd": 950,
|
||||
"release_year": 2023,
|
||||
"description": "中端工作站CPU",
|
||||
"subcategory_id": "server"
|
||||
},
|
||||
{
|
||||
"id": "ryzen97950x",
|
||||
"name": "AMD Ryzen 9 7950X",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 4",
|
||||
"cores": 16,
|
||||
"threads": 32,
|
||||
"base_clock_ghz": 4.5,
|
||||
"boost_clock_ghz": 5.7,
|
||||
"l3_cache_mb": 64,
|
||||
"tdp_watts": 170,
|
||||
"price_usd": 550,
|
||||
"release_year": 2022,
|
||||
"description": "顶级消费级CPU，适合AI开发",
|
||||
"subcategory_id": "desktop"
|
||||
},
|
||||
{
|
||||
"id": "ryzen97950x3d",
|
||||
"name": "AMD Ryzen 9 7950X3D",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 4",
|
||||
"cores": 16,
|
||||
"threads": 32,
|
||||
"base_clock_ghz": 4.2,
|
||||
"boost_clock_ghz": 5.7,
|
||||
"l3_cache_mb": 144,
|
||||
"tdp_watts": 120,
|
||||
"price_usd": 700,
|
||||
"release_year": 2023,
|
||||
"description": "带3D V-Cache，游戏性能更强",
|
||||
"subcategory_id": "mobile"
|
||||
},
|
||||
{
|
||||
"id": "intel14900k",
|
||||
"name": "Intel Core i9-14900K",
|
||||
"manufacturer": "Intel",
|
||||
"architecture": "Raptor Lake Refresh",
|
||||
"cores": 24,
|
||||
"threads": 32,
|
||||
"base_clock_ghz": 3.2,
|
||||
"boost_clock_ghz": 6.0,
|
||||
"l3_cache_mb": 36,
|
||||
"tdp_watts": 125,
|
||||
"price_usd": 580,
|
||||
"release_year": 2023,
|
||||
"description": "Intel顶级消费级CPU",
|
||||
"subcategory_id": "desktop"
|
||||
},
|
||||
{
|
||||
"name": "AMD 锐龙 AI 9 H 365",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "Zen 5, Zen 5c",
|
||||
"cores": 10,
|
||||
"threads": 20,
|
||||
"base_clock_ghz": 2.0,
|
||||
"boost_clock_ghz": 5.0,
|
||||
"l3_cache_mb": 24,
|
||||
"tdp_watts": 28,
|
||||
"description": "AMD 锐龙 AI 处理器助力打造卓越 AI PC",
|
||||
"id": "52af6cf2dc28",
|
||||
"created_at": "2026-04-20 23:19:20",
|
||||
"visible": true,
|
||||
"raw_text": "AMD 锐龙 AI 9 H 365\nAMD 锐龙 AI 处理器助力打造卓越 AI PC\n\n \n全部折叠\n一般规格\n名称\nAMD 锐龙 AI 9 H 365\n产品系列\n锐龙\n系列\n锐龙 AI 300 系列\n外形规格\n笔记本电脑 , 台式机\nAMD PRO 技术\n否\n区域供货状况\n中国\n原代号\nStrix Point\n处理器架构\n4x Zen 5 , 6x Zen 5c\nCPU 核心数\n10\n多线程 (SMT)\n是\n线程数\n20\n最高加速时钟频率 \n最高可达 5 GHz\nMax Zen5c Clock \n最高可达 3.3 GHz\n基准时钟频率 \n2 GHz\nZen5 Base Clock\n2 GHz\nZen5c Base Clock\n2 GHz\nL2 高速缓存\n10 MB\nL3 高速缓存\n24 MB\n默认热设计功耗 (TDP)\n28W\nAMD 可配置热设计功耗 (cTDP)\n15-54W\nCPU 核心的处理器工艺\nTSMC 4nm FinFET\n封装芯片计数\n1\nAMD EXPO™ 内存超频技术\n是\n精准频率提升 (PBO)\n是\n曲线优化器电压偏移\n是\nCPU 平台\nFP8\n支持的扩展\nAES , AMD-V , AVX , AVX2 , AVX512 , FMA3 , MMX-plus , SHA , SSE , SSE2 , SSE3 , SSE4.1 , SSE4.2 , SSE4A , SSSE3 , x86-64\n最高工作温度 (Tjmax)\n100°C\n*支持的操作系统\nWindows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit\n连接\nNative USB 4 (40Gbps)\n2\nNative USB 3.2 Gen 2 (10Gbps)\n2\nNative USB 2.0 (480Mbps)\n4\nPCI Express® Version\nPCIe® 4.0\n原生 PCIe® 通道 (总共/可用)\n16 , 16\nNVMe 支持\nBoot , RAID0 , RAID1\n系统内存类型\nDDR5 (FP8) , LPDDR5X (FP8)\n内存通道数\n2\n最大内存\n256 GB\n最高内存速度\n2x2R\tDDR5-5600, LPDDR5x-8000\n支持 ECC\n否\n显卡功能\n显卡型号\nAMD Radeon™ 880M\n显卡核心数\n12\n显卡频率\n2900 MHz\nDirectX® 版本\n12\nDisplayPort™ 版本\n2.1\nDisplayPort 扩展功能\nAdaptive-Sync , HDR Metadata , UHBR10\nDisplayPort 最高刷新率 (SDR)\n7680x4320 @ 60Hz , 3840x2160 @ 240Hz , 3440x1440 @ 360Hz , 2560x1440 @ 480Hz , 1920x1080 @ 600Hz\nDisplayPort 最高刷新率 (HDR)\n7680x4320 @ 60Hz , 3840x2160 @ 240Hz , 3440x1440 @ 360Hz , 2560x1440 @ 480Hz , 1920x1080 @ 600Hz\nHDMI® 版本\n2.1\n支持的 HDCP 版本\n2.3\nUSB Type-C® DisplayPort™ 备用模式\n是\n支持多个显示器\n是\n显示器个数上限\n4\nAMD FreeSync™\n是\n无线显示\nMiracast\n最大视频编码带宽 (SDR)\n1080p630 8bpc H.264, 1440p373 8bpc H.264, 2160p175 8bpc H.264, 1080p630 8bpc H.265, 1440p373 8bpc H.265, 2160p175 8bpc H.265, 4320p43 8bpc H.265, 1080p864 8/10bpc AV1, 1440p513 8/10bpc AV1, 2160p240 8/10bpc AV1, 4320p60 8/10bpc AV1\n\n最大视频解码带宽\n1080p60 8bpc MPEG2, 1080p60 8bpc VC1, 1080p786 8/10bpc VP9, 2160p196 8/10bpc VP9, 4320p49 8/10bpc VP9, 1080p1200 8bpc H.264, 2160p300 8bpc H.264, 4320p75 8bpc H.264, 1080p786 8/10bpc H.265, 2160p196 8/10bpc H.265, 4320p49 8/10bpc H.265, 1080p960 8/10bpc\n\nAMD SmartShift MAX\n是\nAMD 显存智取技术\n支持\nAI 引擎性能\nAMD Ryzen™ AI\n支持\nOverall TOPS\n最高可达 73 TOPS\nNPU TOPS\n最高可达 50 TOPS\n产品 ID\nTray 产品 ID\n100-000001530 (FP8)\n安全\nAMD 增强病毒防护 (NX bit)\n是",
|
||||
"publish_date": "",
|
||||
"views": 0,
|
||||
"is_pinned": false,
|
||||
"subcategory_id": "mobile"
|
||||
}
|
||||
]
|
||||
215
data/gpus.json
215
data/gpus.json
@@ -1,12 +1,207 @@
|
||||
[
|
||||
{"id": "h100", "name": "NVIDIA H100", "manufacturer": "NVIDIA", "architecture": "Hopper", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbs": 3352, "fp32_tflops": 67, "fp16_tflops": 1979, "int8_perf_tops": 3958, "price_usd": 30000, "release_year": 2022, "description": "数据中心顶级GPU，专为AI训练设计"},
|
||||
{"id": "a100", "name": "NVIDIA A100", "manufacturer": "NVIDIA", "architecture": "Ampere", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbs": 2039, "fp32_tflops": 19.5, "fp16_tflops": 312, "int8_perf_tops": 624, "price_usd": 10000, "release_year": 2020, "description": "数据中心主力GPU，AI训练推理通用"},
|
||||
{"id": "a10040g", "name": "NVIDIA A100 40GB", "manufacturer": "NVIDIA", "architecture": "Ampere", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbs": 1555, "fp32_tflops": 19.5, "fp16_tflops": 312, "int8_perf_tops": 624, "price_usd": 6000, "release_year": 2020, "description": "A100 40GB版本，性价比更高"},
|
||||
{"id": "l40s", "name": "NVIDIA L40S", "manufacturer": "NVIDIA", "architecture": "Ada Lovelace", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbs": 864, "fp32_tflops": 91.6, "fp16_tflops": 362, "int8_perf_tops": 724, "price_usd": 7000, "release_year": 2023, "description": "新一代数据中心GPU，推理优化"},
|
||||
{"id": "rtx4090", "name": "NVIDIA RTX 4090", "manufacturer": "NVIDIA", "architecture": "Ada Lovelace", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbs": 1008, "fp32_tflops": 82.6, "fp16_tflops": 330, "int8_perf_tops": 660, "price_usd": 1600, "release_year": 2022, "description": "消费级最强GPU，适合个人AI开发"},
|
||||
{"id": "rtx4090d", "name": "NVIDIA RTX 4090D", "manufacturer": "NVIDIA", "architecture": "Ada Lovelace", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 24, "memory_bandwidth_gbs": 1008, "fp32_tflops": 73.5, "fp16_tflops": 294, "int8_perf_tops": 588, "price_usd": 1400, "release_year": 2024, "description": "4090中国特供版，性能略降"},
|
||||
{"id": "rtx3090", "name": "NVIDIA RTX 3090", "manufacturer": "NVIDIA", "architecture": "Ampere", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbs": 936, "fp32_tflops": 35.6, "fp16_tflops": 142, "int8_perf_tops": 284, "price_usd": 1200, "release_year": 2020, "description": "上一代旗舰，性价比高"},
|
||||
{"id": "rtx3080", "name": "NVIDIA RTX 3080", "manufacturer": "NVIDIA", "architecture": "Ampere", "cuda_cores": 8704, "tensor_cores": 272, "memory_gb": 10, "memory_bandwidth_gbs": 760, "fp32_tflops": 29.8, "fp16_tflops": 119, "int8_perf_tops": 238, "price_usd": 700, "release_year": 2020, "description": "中高端消费级GPU"},
|
||||
{"id": "v100", "name": "NVIDIA V100", "manufacturer": "NVIDIA", "architecture": "Volta", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbs": 900, "fp32_tflops": 14.8, "fp16_tflops": 118, "int8_perf_tops": 236, "price_usd": 4000, "release_year": 2017, "description": "上一代数据中心GPU，仍有价值"},
|
||||
{"id": "mi300x", "name": "AMD MI300X", "manufacturer": "AMD", "architecture": "CDNA 3", "cuda_cores": 0, "tensor_cores": 304, "memory_gb": 192, "memory_bandwidth_gbs": 5300, "fp32_tflops": 81.7, "fp16_tflops": 1307, "int8_perf_tops": 2614, "price_usd": 15000, "release_year": 2023, "description": "AMD最强AI GPU，192GB显存"}
|
||||
{
|
||||
"id": "h100",
|
||||
"name": "NVIDIA H100",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Hopper",
|
||||
"cuda_cores": 16896,
|
||||
"tensor_cores": 528,
|
||||
"memory_gb": 80,
|
||||
"memory_bandwidth_gbs": 3352,
|
||||
"fp32_tflops": 67,
|
||||
"fp16_tflops": 1979,
|
||||
"int8_perf_tops": 3958,
|
||||
"price_usd": 30000,
|
||||
"release_year": 2022,
|
||||
"description": "数据中心顶级GPU，专为AI训练设计",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"id": "a100",
|
||||
"name": "NVIDIA A100",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ampere",
|
||||
"cuda_cores": 6912,
|
||||
"tensor_cores": 432,
|
||||
"memory_gb": 80,
|
||||
"memory_bandwidth_gbs": 2039,
|
||||
"fp32_tflops": 19.5,
|
||||
"fp16_tflops": 312,
|
||||
"int8_perf_tops": 624,
|
||||
"price_usd": 10000,
|
||||
"release_year": 2020,
|
||||
"description": "数据中心主力GPU，AI训练推理通用",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"id": "a10040g",
|
||||
"name": "NVIDIA A100 40GB",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ampere",
|
||||
"cuda_cores": 6912,
|
||||
"tensor_cores": 432,
|
||||
"memory_gb": 40,
|
||||
"memory_bandwidth_gbs": 1555,
|
||||
"fp32_tflops": 19.5,
|
||||
"fp16_tflops": 312,
|
||||
"int8_perf_tops": 624,
|
||||
"price_usd": 6000,
|
||||
"release_year": 2020,
|
||||
"description": "A100 40GB版本，性价比更高",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"id": "l40s",
|
||||
"name": "NVIDIA L40S",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ada Lovelace",
|
||||
"cuda_cores": 18176,
|
||||
"tensor_cores": 568,
|
||||
"memory_gb": 48,
|
||||
"memory_bandwidth_gbs": 864,
|
||||
"fp32_tflops": 91.6,
|
||||
"fp16_tflops": 362,
|
||||
"int8_perf_tops": 724,
|
||||
"price_usd": 7000,
|
||||
"release_year": 2023,
|
||||
"description": "新一代数据中心GPU，推理优化",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"id": "rtx4090",
|
||||
"name": "NVIDIA RTX 4090",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ada Lovelace",
|
||||
"cuda_cores": 16384,
|
||||
"tensor_cores": 512,
|
||||
"memory_gb": 24,
|
||||
"memory_bandwidth_gbs": 1008,
|
||||
"fp32_tflops": 82.6,
|
||||
"fp16_tflops": 330,
|
||||
"int8_perf_tops": 660,
|
||||
"price_usd": 1600,
|
||||
"release_year": 2022,
|
||||
"description": "消费级最强GPU，适合个人AI开发",
|
||||
"subcategory_id": "gaming"
|
||||
},
|
||||
{
|
||||
"id": "rtx4090d",
|
||||
"name": "NVIDIA RTX 4090D",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ada Lovelace",
|
||||
"cuda_cores": 14592,
|
||||
"tensor_cores": 456,
|
||||
"memory_gb": 24,
|
||||
"memory_bandwidth_gbs": 1008,
|
||||
"fp32_tflops": 73.5,
|
||||
"fp16_tflops": 294,
|
||||
"int8_perf_tops": 588,
|
||||
"price_usd": 1400,
|
||||
"release_year": 2024,
|
||||
"description": "4090中国特供版，性能略降",
|
||||
"subcategory_id": "gaming"
|
||||
},
|
||||
{
|
||||
"id": "rtx3090",
|
||||
"name": "NVIDIA RTX 3090",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ampere",
|
||||
"cuda_cores": 10496,
|
||||
"tensor_cores": 328,
|
||||
"memory_gb": 24,
|
||||
"memory_bandwidth_gbs": 936,
|
||||
"fp32_tflops": 35.6,
|
||||
"fp16_tflops": 142,
|
||||
"int8_perf_tops": 284,
|
||||
"price_usd": 1200,
|
||||
"release_year": 2020,
|
||||
"description": "上一代旗舰，性价比高",
|
||||
"subcategory_id": "gaming"
|
||||
},
|
||||
{
|
||||
"id": "rtx3080",
|
||||
"name": "NVIDIA RTX 3080",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Ampere",
|
||||
"cuda_cores": 8704,
|
||||
"tensor_cores": 272,
|
||||
"memory_gb": 10,
|
||||
"memory_bandwidth_gbs": 760,
|
||||
"fp32_tflops": 29.8,
|
||||
"fp16_tflops": 119,
|
||||
"int8_perf_tops": 238,
|
||||
"price_usd": 700,
|
||||
"release_year": 2020,
|
||||
"description": "中高端消费级GPU",
|
||||
"subcategory_id": "gaming"
|
||||
},
|
||||
{
|
||||
"id": "v100",
|
||||
"name": "NVIDIA V100",
|
||||
"manufacturer": "NVIDIA",
|
||||
"architecture": "Volta",
|
||||
"cuda_cores": 5120,
|
||||
"tensor_cores": 640,
|
||||
"memory_gb": 32,
|
||||
"memory_bandwidth_gbs": 900,
|
||||
"fp32_tflops": 14.8,
|
||||
"fp16_tflops": 118,
|
||||
"int8_perf_tops": 236,
|
||||
"price_usd": 4000,
|
||||
"release_year": 2017,
|
||||
"description": "上一代数据中心GPU，仍有价值",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"id": "mi300x",
|
||||
"name": "AMD MI300X",
|
||||
"manufacturer": "AMD",
|
||||
"architecture": "CDNA 3",
|
||||
"cuda_cores": 0,
|
||||
"tensor_cores": 304,
|
||||
"memory_gb": 192,
|
||||
"memory_bandwidth_gbs": 5300,
|
||||
"fp32_tflops": 81.7,
|
||||
"fp16_tflops": 1307,
|
||||
"int8_perf_tops": 2614,
|
||||
"price_usd": 15000,
|
||||
"release_year": 2023,
|
||||
"description": "AMD最强AI GPU，192GB显存",
|
||||
"subcategory_id": "datacenter"
|
||||
},
|
||||
{
|
||||
"name": "RTX 6000D",
|
||||
"manufacturer": "NVIDIA",
|
||||
"memory_gb": 84,
|
||||
"cuda_cores": 19968,
|
||||
"description": "NVIDIA为中国市场定制的全新工作站显卡，搭载84GB GDDR7显存、19968个CUDA核心，采用被动散热设计，专为服务器机箱风道优化。显存总线为448位，核心频率为2430MHz，在Geekbench 6 OpenCL测试中获得390,656分。",
|
||||
"id": "f56b2de6fac4",
|
||||
"created_at": "2026-04-20 18:19:14",
|
||||
"visible": true,
|
||||
"raw_text": "据tweaktown报道，NVIDIA为中国市场定制的全新工作站显卡RTX 6000D近日迎来首度拆解。该卡搭载84GB GDDR7显存、19968个CUDA核心，采用被动散热设计，专为服务器机箱风道优化。\n\n\n相较于满血RTX PRO 6000（96GB GDDR7/512-bit），中国特供版RTX 6000D在规格上进行了多处调整。国内团队“技数犬”发布了拆解视频。\n\n据了解，RTX 6000D为无风扇被动散热设计，完全依靠机箱气流降温。\n\nRTX 6000D搭载28颗VRAM模块，总计84GB GDDR7显存，显存总线为448位（相比RTX PRO 6000的96GB/512位有所减少）。\n\nRTX 6000D GPU 核心为156 SM单元，19,968个CUDA核心，比RTX PRO 6000少约17%。\n\nRTX 6000D核心频率为2430MHz（RTX PRO 6000为2600MHz），TDP暂未公布。性能方面，RTX 6000D在Geekbench 6 OpenCL测试中获得390,656分，低于RTX PRO 6000的45–50万分。",
|
||||
"currency": "CNY",
|
||||
"price_usd": 45000,
|
||||
"updated_at": "2026-04-28 11:56:48",
|
||||
"subcategory_id": "professional",
|
||||
"views": 0,
|
||||
"images": []
|
||||
},
|
||||
{
|
||||
"name": "RTX PRO 6000",
|
||||
"description": "这款专业显卡基于 GB202 GPU，拥有 24064 个 CUDA 核心（188 个 SM），运行频率达 2,617 MHz，并配备 96 GB 支持 ECC 校验的 GDDR7 显存。\n\n相比之下，面向游戏市场的旗舰显卡 RTX 5090 虽同样基于 GB202 ，但其 CUDA 核心数量缩减至 21,760 个，频率为 2,410 MHz，显存容量为 32 GB。\n\n96G超大显存RTX PRO 6000Blackwell初次跑分，略逊于RTX 5090\n其测试平台采用了华硕 Pro WS WRX80E-SAGE SE WIFI 主板、AMD 锐龙 Threadripper PRO 3975WX 处理器、512 GB 内存。\n\n在 Geekbench 6.4.0 上，其测试平台 OpenCL 得分仅 368219 分，略低于 RTX 5090 的 376,858 分，差距约 2.3%，外媒认为这主要是由于 RTX PRO 6000 缺乏正式版驱动导致，且显卡功耗可能受限。\n\nRTX PRO 6000 系列将提供两种版本，分别为适用于紧凑型机箱，规格相同的Max-Q 工作站版，但TDP 功耗限制在 300W；以及支持最高600W TDP的标准版，可满足高强度计算需求。",
|
||||
"id": "d246301f2032",
|
||||
"created_at": "2026-04-20 18:21:00",
|
||||
"visible": true,
|
||||
"raw_text": "这款专业显卡基于 GB202 GPU，拥有 24064 个 CUDA 核心（188 个 SM），运行频率达 2,617 MHz，并配备 96 GB 支持 ECC 校验的 GDDR7 显存。\n\n相比之下，面向游戏市场的旗舰显卡 RTX 5090 虽同样基于 GB202 ，但其 CUDA 核心数量缩减至 21,760 个，频率为 2,410 MHz，显存容量为 32 GB。\n\n96G超大显存RTX PRO 6000Blackwell初次跑分，略逊于RTX 5090\n其测试平台采用了华硕 Pro WS WRX80E-SAGE SE WIFI 主板、AMD 锐龙 Threadripper PRO 3975WX 处理器、512 GB 内存。\n\n在 Geekbench 6.4.0 上，其测试平台 OpenCL 得分仅 368219 分，略低于 RTX 5090 的 376,858 分，差距约 2.3%，外媒认为这主要是由于 RTX PRO 6000 缺乏正式版驱动导致，且显卡功耗可能受限。\n\nRTX PRO 6000 系列将提供两种版本，分别为适用于紧凑型机箱，规格相同的Max-Q 工作站版，但TDP 功耗限制在 300W；以及支持最高600W TDP的标准版，可满足高强度计算需求。",
|
||||
"architecture": "GB202",
|
||||
"memory_gb": 96,
|
||||
"cuda_cores": 24064,
|
||||
"currency": "CNY",
|
||||
"price_usd": 65000,
|
||||
"updated_at": "2026-04-28 11:56:38",
|
||||
"manufacturer": "NVIDIA",
|
||||
"subcategory_id": "professional",
|
||||
"views": 0,
|
||||
"images": []
|
||||
}
|
||||
]
|
||||
@@ -2,10 +2,38 @@
|
||||
{
|
||||
"name": "比亚迪宋plus dmi 2021款",
|
||||
"brand": "比亚迪",
|
||||
"price": "18.87",
|
||||
"price": 18.87,
|
||||
"year": "2021",
|
||||
"category_id": "021dc76d36be",
|
||||
"id": "3d20dbcd4bdd",
|
||||
"created_at": "2026-04-09 10:09:56"
|
||||
"created_at": "2026-04-09 10:09:56",
|
||||
"subcategory_id": "suv",
|
||||
"views": 0,
|
||||
"images": [],
|
||||
"updated_at": "2026-04-28 12:32:13"
|
||||
},
|
||||
{
|
||||
"name": "秦PLUS",
|
||||
"brand": "比亚迪",
|
||||
"price": 5.98,
|
||||
"specs": {
|
||||
"长宽高": "4780*1837*1515mm",
|
||||
"轴距": "2718mm",
|
||||
"前轮距": "1580mm",
|
||||
"后轮距": "1590mm",
|
||||
"轮胎规格": "225/60 R16",
|
||||
"发动机": "1.5L 101马力 L4",
|
||||
"最大功率": "74kW",
|
||||
"最大扭矩": "126N·m",
|
||||
"变速器": "E-CVT无级变速器",
|
||||
"中控屏幕尺寸": "10.1英寸"
|
||||
},
|
||||
"description": "秦PLUS是一款外观设计极具现代感和运动气息的车型，采用家族化设计语言，配备大尺寸进气格栅和锐利LED大灯。车身线条流畅，内饰简洁大气，配备10.1英寸中控屏和语音识别系统。搭载1.5L发动机和E-CVT变速器，提供平稳动力输出和低油耗。内部空间宽敞，座椅舒适，支持多种调节功能。",
|
||||
"id": "b78fc7983a70",
|
||||
"category_id": "021dc76d36be",
|
||||
"created_at": "2026-04-11 02:03:45",
|
||||
"visible": true,
|
||||
"raw_text": "秦PLUS的外观设计极具现代感和运动气息，前脸采用了家族化设计语言，标志性的大尺寸进气格栅占据了前脸的大部分空间，搭配锐利的LED大灯组，营造出强烈的视觉冲击力。车身线条流畅，腰线从车头贯穿至车尾，增强了整车的运动感。车尾部分，简洁大方的设计与前脸相呼应，整体风格时尚而不失稳重。\n\n上海：秦PLUS优惠促销，最新报价5.98万！轻松开新车\n\n秦PLUS拥有4780*1837*1515mm的长宽高尺寸和2718mm的轴距，赋予其宽敞的内部空间。车侧线条流畅且动感十足，从前轮距1580mm到后轮距1590mm，车轮布局合理，增强了车辆的稳定性和操控性。配备的225/60 R16轮胎规格，匹配独特风格的轮圈，为车辆增添了一抹动感与时尚的气息。\n\n上海：秦PLUS优惠促销，最新报价5.98万！轻松开新车\n\n秦PLUS的内饰风格简洁大气，给人以科技感和舒适感。中控台布局合理，配备了10.1英寸的中控屏幕，支持语音识别控制系统，可轻松操作多媒体系统、导航、电话和空调等功能。方向盘采用皮质材料，手感舒适，支持手动上下和前后调节，方便驾驶员调整到最佳驾驶姿势。座椅采用仿皮材质，主驾驶座椅具备前后调节、靠背调节和高低调节功能，而副驾驶座椅则支持前后调节和靠背调节，确保了乘客的舒适度。后排座椅可以按比例放倒，增加储物空间，同时，车内还配备了USB和Type-C接口，方便乘客为电子设备充电。\n\n上海：秦PLUS优惠促销，最新报价5.98万！轻松开新车\n\n秦PLUS搭载了一台1.5L 101马力的L4发动机，最大功率为74kW，最大扭矩为126N·m。与之匹配的是E-CVT无级变速器，这使得车辆在提供平稳的动力输出的同时，还能有效降低油耗。\n\n汽车之家车主@天艺风云 表示，外观设计是他当初选择秦PLUS的原因之一。他赞赏整体造型时尚大气，龙脸设计搭配犀利的大灯，辨识度极高。车身线条流畅，溜背式造型增添了几分运动感。全新的“龙鳞辉熠”格栅，精致又霸气，每次停车都有人问这是什么车，外观确实很吸引人。",
|
||||
"subcategory_id": "sedan"
|
||||
}
|
||||
]
|
||||
53
data/items_71fa2b4d818f.json
Normal file
53
data/items_71fa2b4d818f.json
Normal file
@@ -0,0 +1,53 @@
|
||||
[
|
||||
{
|
||||
"name": "Osmo Pocket 4",
|
||||
"brand": "DJI",
|
||||
"price": 2999,
|
||||
"specs": "[object Object]",
|
||||
"id": "597e29af5937",
|
||||
"category_id": "71fa2b4d818f",
|
||||
"created_at": "2026-04-28 00:07:01",
|
||||
"visible": true,
|
||||
"raw_text": "",
|
||||
"images": [],
|
||||
"publish_date": "",
|
||||
"views": 0,
|
||||
"is_pinned": false,
|
||||
"subcategory_id": "90ce312b560d",
|
||||
"updated_at": "2026-04-28 12:32:38"
|
||||
},
|
||||
{
|
||||
"name": "Osmo Pocket 3",
|
||||
"brand": "DJI",
|
||||
"price": 2799,
|
||||
"specs": "[object Object]",
|
||||
"id": "ad10ac80827b",
|
||||
"category_id": "71fa2b4d818f",
|
||||
"created_at": "2026-04-28 00:07:01",
|
||||
"visible": true,
|
||||
"raw_text": "",
|
||||
"images": [],
|
||||
"publish_date": "",
|
||||
"views": 0,
|
||||
"is_pinned": false,
|
||||
"subcategory_id": "90ce312b560d",
|
||||
"updated_at": "2026-04-28 12:32:43"
|
||||
},
|
||||
{
|
||||
"name": "DJI Pocket 2",
|
||||
"brand": "DJI",
|
||||
"price": 1999,
|
||||
"specs": "[object Object]",
|
||||
"id": "0fde0f10ad96",
|
||||
"category_id": "71fa2b4d818f",
|
||||
"created_at": "2026-04-28 00:07:01",
|
||||
"visible": true,
|
||||
"raw_text": "",
|
||||
"images": [],
|
||||
"publish_date": "",
|
||||
"views": 0,
|
||||
"is_pinned": false,
|
||||
"subcategory_id": "90ce312b560d",
|
||||
"updated_at": "2026-04-28 12:32:50"
|
||||
}
|
||||
]
|
||||
@@ -1,9 +1,66 @@
|
||||
[
|
||||
{"id": "k001", "title": "什么是参数量？", "category": "ai-models", "icon": "ri-calculator-line", "content": "参数量（Parameters）是衡量大模型规模的指标，表示模型中权重参数的数量。例如 GPT-3 有 175B 参数，即约1750亿个参数。", "detail": "参数量决定了模型的容量和表达能力。一般来说，参数量越大，模型能力越强，但也需要更多计算资源。\n\n常见规模分类：\n- 小模型：<1B (适合边缘设备)\n- 中模型：1B-10B (消费级GPU可运行)\n- 大模型：10B-100B (需要多GPU)\n- 超大模型：>100B (需要数据中心)", "order": 1},
|
||||
{"id": "k002", "title": "什么是上下文长度？", "category": "ai-models", "icon": "ri-text-wrap", "content": "上下文长度（Context Length）是模型能处理的输入文本最大长度。更长的上下文意味着模型可以理解更长的文档或对话历史。", "detail": "常见长度：\n- 4K：传统长度，适合简单对话\n- 32K：中等长度，适合长文档\n- 128K：超长上下文，如GPT-4 Turbo\n- 200K：Claude 3的极限长度", "order": 2},
|
||||
{"id": "k003", "title": "什么是量化？", "category": "ai-models", "icon": "ri-scales-3-line", "content": "量化（Quantization）是将模型参数从高精度转换为低精度，减少显存占用和计算量。如FP16→INT8→INT4，精度损失可控，资源节省显著。", "detail": "量化效果：\n- FP32→FP16: 显存减半，精度基本不变\n- FP16→INT8: 显存再减半，精度略降\n- INT8→INT4: 显存再减半，需特殊技术\n\n推荐工具：llama.cpp、GPTQ、AWQ等", "order": 3},
|
||||
{"id": "k004", "title": "什么是MMLU？", "category": "ai-models", "icon": "ri-bar-chart-box-line", "content": "MMLU（Massive Multitask Language Understanding）是评估大模型综合能力的标准测试集，覆盖57个学科领域。", "detail": "分数参考：\n- 60-70%：入门级，如GPT-3\n- 70-80%：中等水平，如Llama 2 70B\n- 80-90%：优秀水平，如GPT-4、Claude 3", "order": 4},
|
||||
{"id": "k005", "title": "如何计算显存需求？", "category": "gpus", "icon": "ri-memory-line", "content": "模型显存需求 ≈ 参数量 × 每参数字节数 × 1.3（含KV Cache开销）", "detail": "计算公式：\n- FP32: 参数量 × 4字节 × 1.3\n- FP16: 参数量 × 2字节 × 1.3\n- INT8: 参数量 × 1字节 × 1.3\n- INT4: 参数量 × 0.5字节 × 1.3\n\n例如：7B模型FP16加载需要约 7 × 2 × 1.3 ≈ 18GB显存", "order": 1},
|
||||
{"id": "k006", "title": "GPU架构演进", "category": "gpus", "icon": "ri-history-line", "content": "NVIDIA GPU架构从Fermi到Hopper，每一代都有显著提升。了解架构有助于选择合适的GPU。", "detail": "主要架构：\n- Volta (2017): V100, 引入Tensor Core\n- Turing (2018): RTX 20系列, RT Core\n- Ampere (2020): A100, RTX 30系列\n- Hopper (2022): H100, FP8支持\n- Ada Lovelace (2022): RTX 40系列, L40S", "order": 2},
|
||||
{"id": "k007", "title": "CPU核心数选择", "category": "cpus", "icon": "ri-database-2-line", "content": "CPU核心数的选择取决于应用场景。更多核心适合并行任务，但单核性能也很重要。", "detail": "场景推荐：\n- 办公/日常：4-6核足够\n- 开发/编译：8-16核\n- 服务器/虚拟化：16-64核\n- 高性能计算：64核以上\n\n注意：AI训练主要依赖GPU，CPU主要用于数据预处理", "order": 1}
|
||||
{
|
||||
"id": "k001",
|
||||
"title": "什么是参数量？",
|
||||
"category": "ai-models",
|
||||
"icon": "ri-calculator-line",
|
||||
"content": "参数量（Parameters）是衡量大模型规模的指标，表示模型中权重参数的数量。例如 GPT-3 有 175B 参数，即约1750亿个参数。",
|
||||
"detail": "参数量决定了模型的容量和表达能力。一般来说，参数量越大，模型能力越强，但也需要更多计算资源。\n\n常见规模分类：\n- 小模型：<1B (适合边缘设备)\n- 中模型：1B-10B (消费级GPU可运行)\n- 大模型：10B-100B (需要多GPU)\n- 超大模型：>100B (需要数据中心)",
|
||||
"order": 1
|
||||
},
|
||||
{
|
||||
"id": "k002",
|
||||
"title": "什么是上下文长度？",
|
||||
"category": "ai-models",
|
||||
"icon": "ri-text-wrap",
|
||||
"content": "上下文长度（Context Length）是模型能处理的输入文本最大长度。更长的上下文意味着模型可以理解更长的文档或对话历史。",
|
||||
"detail": "常见长度：\n- 4K：传统长度，适合简单对话\n- 32K：中等长度，适合长文档\n- 128K：超长上下文，如GPT-4 Turbo\n- 200K：Claude 3的极限长度",
|
||||
"order": 2
|
||||
},
|
||||
{
|
||||
"id": "k003",
|
||||
"title": "什么是量化？",
|
||||
"category": "ai-models",
|
||||
"icon": "ri-scales-3-line",
|
||||
"content": "量化（Quantization）是将模型参数从高精度转换为低精度，减少显存占用和计算量。如FP16→INT8→INT4，精度损失可控，资源节省显著。",
|
||||
"detail": "量化效果：\n- FP32→FP16: 显存减半，精度基本不变\n- FP16→INT8: 显存再减半，精度略降\n- INT8→INT4: 显存再减半，需特殊技术\n\n推荐工具：llama.cpp、GPTQ、AWQ等",
|
||||
"order": 3
|
||||
},
|
||||
{
|
||||
"id": "k004",
|
||||
"title": "什么是MMLU？",
|
||||
"category": "ai-models",
|
||||
"icon": "ri-bar-chart-box-line",
|
||||
"content": "MMLU（Massive Multitask Language Understanding）是评估大模型综合能力的标准测试集，覆盖57个学科领域。",
|
||||
"detail": "分数参考：\n- 60-70%：入门级，如GPT-3\n- 70-80%：中等水平，如Llama 2 70B\n- 80-90%：优秀水平，如GPT-4、Claude 3",
|
||||
"order": 4,
|
||||
"visible": false
|
||||
},
|
||||
{
|
||||
"id": "k005",
|
||||
"title": "如何计算显存需求？",
|
||||
"category": "gpus",
|
||||
"icon": "ri-memory-line",
|
||||
"content": "模型显存需求 ≈ 参数量 × 每参数字节数 × 1.3（含KV Cache开销）",
|
||||
"detail": "计算公式：\n- FP32: 参数量 × 4字节 × 1.3\n- FP16: 参数量 × 2字节 × 1.3\n- INT8: 参数量 × 1字节 × 1.3\n- INT4: 参数量 × 0.5字节 × 1.3\n\n例如：7B模型FP16加载需要约 7 × 2 × 1.3 ≈ 18GB显存",
|
||||
"order": 1
|
||||
},
|
||||
{
|
||||
"id": "k006",
|
||||
"title": "GPU架构演进",
|
||||
"category": "gpus",
|
||||
"icon": "ri-history-line",
|
||||
"content": "NVIDIA GPU架构从Fermi到Hopper，每一代都有显著提升。了解架构有助于选择合适的GPU。",
|
||||
"detail": "主要架构：\n- Volta (2017): V100, 引入Tensor Core\n- Turing (2018): RTX 20系列, RT Core\n- Ampere (2020): A100, RTX 30系列\n- Hopper (2022): H100, FP8支持\n- Ada Lovelace (2022): RTX 40系列, L40S",
|
||||
"order": 2
|
||||
},
|
||||
{
|
||||
"id": "k007",
|
||||
"title": "CPU核心数选择",
|
||||
"category": "cpus",
|
||||
"icon": "ri-database-2-line",
|
||||
"content": "CPU核心数的选择取决于应用场景。更多核心适合并行任务，但单核性能也很重要。",
|
||||
"detail": "场景推荐：\n- 办公/日常：4-6核足够\n- 开发/编译：8-16核\n- 服务器/虚拟化：16-64核\n- 高性能计算：64核以上\n\n注意：AI训练主要依赖GPU，CPU主要用于数据预处理",
|
||||
"order": 1
|
||||
}
|
||||
]
|
||||
221
data/models.json
221
data/models.json
@@ -1,14 +1,211 @@
|
||||
[
|
||||
{"id": "gpt4", "name": "GPT-4", "organization": "OpenAI", "parameters": 1760, "architecture": "Transformer", "context_length": 8192, "input_price": 0.03, "output_price": 0.06, "mmlu": 86.4, "humaneval": 67.0, "is_open_source": false, "license": "Proprietary", "description": "OpenAI最强大的多模态大模型", "created_at": "2024-01-01"},
|
||||
{"id": "gpt4turbo", "name": "GPT-4 Turbo", "organization": "OpenAI", "parameters": 1760, "architecture": "Transformer", "context_length": 128000, "input_price": 0.01, "output_price": 0.03, "mmlu": 86.4, "humaneval": 67.0, "is_open_source": false, "license": "Proprietary", "description": "GPT-4增强版，128K上下文", "created_at": "2024-01-01"},
|
||||
{"id": "gpt35", "name": "GPT-3.5 Turbo", "organization": "OpenAI", "parameters": 175, "architecture": "Transformer", "context_length": 16385, "input_price": 0.0005, "output_price": 0.0015, "mmlu": 70.0, "humaneval": 48.1, "is_open_source": false, "license": "Proprietary", "description": "性价比高的通用模型", "created_at": "2024-01-01"},
|
||||
{"id": "claude3opus", "name": "Claude 3 Opus", "organization": "Anthropic", "parameters": 400, "architecture": "Transformer", "context_length": 200000, "input_price": 0.015, "output_price": 0.075, "mmlu": 86.8, "humaneval": 84.9, "is_open_source": false, "license": "Proprietary", "description": "Anthropic最强模型，200K上下文", "created_at": "2024-01-01"},
|
||||
{"id": "claude3sonnet", "name": "Claude 3 Sonnet", "organization": "Anthropic", "parameters": 175, "architecture": "Transformer", "context_length": 200000, "input_price": 0.003, "output_price": 0.015, "mmlu": 79.0, "humaneval": 73.0, "is_open_source": false, "license": "Proprietary", "description": "平衡性能与成本", "created_at": "2024-01-01"},
|
||||
{"id": "llama270b", "name": "Llama 2 70B", "organization": "Meta", "parameters": 70, "architecture": "Transformer", "context_length": 4096, "input_price": 0, "output_price": 0, "mmlu": 69.8, "humaneval": 29.9, "is_open_source": true, "license": "Llama 2 Community", "description": "Meta开源大模型，70B参数", "created_at": "2024-01-01"},
|
||||
{"id": "llama3", "name": "Llama 3 70B", "organization": "Meta", "parameters": 70, "architecture": "Transformer", "context_length": 8192, "input_price": 0, "output_price": 0, "mmlu": 82.0, "humaneval": 81.7, "is_open_source": true, "license": "Llama 3 Community", "description": "Meta最新开源模型，性能接近GPT-4", "created_at": "2024-01-01"},
|
||||
{"id": "mistral7b", "name": "Mistral 7B", "organization": "Mistral AI", "parameters": 7, "architecture": "Transformer", "context_length": 32768, "input_price": 0, "output_price": 0, "mmlu": 62.5, "humaneval": 26.8, "is_open_source": true, "license": "Apache 2.0", "description": "小巧高效的开源模型", "created_at": "2024-01-01"},
|
||||
{"id": "mixtral8x7b", "name": "Mixtral 8x7B", "organization": "Mistral AI", "parameters": 47, "architecture": "MoE", "context_length": 32768, "input_price": 0, "output_price": 0, "mmlu": 70.6, "humaneval": 40.2, "is_open_source": true, "license": "Apache 2.0", "description": "MoE架构，高效推理", "created_at": "2024-01-01"},
|
||||
{"id": "qwen72b", "name": "Qwen 72B", "organization": "Alibaba", "parameters": 72, "architecture": "Transformer", "context_length": 32768, "input_price": 0, "output_price": 0, "mmlu": 83.1, "humaneval": 65.4, "is_open_source": true, "license": "Apache 2.0", "description": "阿里开源大模型，中文能力强", "created_at": "2024-01-01"},
|
||||
{"id": "deepseekv3", "name": "DeepSeek V3", "organization": "DeepSeek", "parameters": 685, "architecture": "MoE", "context_length": 128000, "input_price": 0.00014, "output_price": 0.00028, "mmlu": 88.5, "humaneval": 86.2, "is_open_source": true, "license": "MIT", "description": "DeepSeek最新模型，性价比极高", "created_at": "2024-01-01"},
|
||||
{"id": "glm4", "name": "GLM-4", "organization": "Zhipu AI", "parameters": 130, "architecture": "Transformer", "context_length": 128000, "input_price": 0.014, "output_price": 0.014, "mmlu": 81.0, "humaneval": 70.0, "is_open_source": false, "license": "Proprietary", "description": "智谱AI大模型，中文能力强", "created_at": "2024-01-01"}
|
||||
{
|
||||
"id": "gpt4",
|
||||
"name": "GPT-4",
|
||||
"organization": "OpenAI",
|
||||
"parameters": 1760,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 8192,
|
||||
"input_price": 0.03,
|
||||
"output_price": 0.06,
|
||||
"mmlu": 86.4,
|
||||
"humaneval": 67,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "OpenAI最强大的多模态大模型",
|
||||
"created_at": "2024-01-01",
|
||||
"updated_at": "2026-04-28 11:57:02",
|
||||
"raw_text": "\nGPT-4 Turbo version with 128K context length, price is $10 per 1M input tokens",
|
||||
"subcategory_id": "chat",
|
||||
"views": 0,
|
||||
"images": []
|
||||
},
|
||||
{
|
||||
"id": "gpt4turbo",
|
||||
"name": "GPT-4 Turbo",
|
||||
"organization": "OpenAI",
|
||||
"parameters": 1760,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 128000,
|
||||
"input_price": 0.01,
|
||||
"output_price": 0.03,
|
||||
"mmlu": 86.4,
|
||||
"humaneval": 67.0,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "GPT-4增强版，128K上下文",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "gpt35",
|
||||
"name": "GPT-3.5 Turbo",
|
||||
"organization": "OpenAI",
|
||||
"parameters": 175,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 16385,
|
||||
"input_price": 0.0005,
|
||||
"output_price": 0.0015,
|
||||
"mmlu": 70.0,
|
||||
"humaneval": 48.1,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "性价比高的通用模型",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "claude3opus",
|
||||
"name": "Claude 3 Opus",
|
||||
"organization": "Anthropic",
|
||||
"parameters": 400,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 200000,
|
||||
"input_price": 0.015,
|
||||
"output_price": 0.075,
|
||||
"mmlu": 86.8,
|
||||
"humaneval": 84.9,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "Anthropic最强模型，200K上下文",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "code"
|
||||
},
|
||||
{
|
||||
"id": "claude3sonnet",
|
||||
"name": "Claude 3 Sonnet",
|
||||
"organization": "Anthropic",
|
||||
"parameters": 175,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 200000,
|
||||
"input_price": 0.003,
|
||||
"output_price": 0.015,
|
||||
"mmlu": 79.0,
|
||||
"humaneval": 73.0,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "平衡性能与成本",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "llama270b",
|
||||
"name": "Llama 2 70B",
|
||||
"organization": "Meta",
|
||||
"parameters": 70,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 4096,
|
||||
"input_price": 0,
|
||||
"output_price": 0,
|
||||
"mmlu": 69.8,
|
||||
"humaneval": 29.9,
|
||||
"is_open_source": true,
|
||||
"license": "Llama 2 Community",
|
||||
"description": "Meta开源大模型，70B参数",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "llama3",
|
||||
"name": "Llama 3 70B",
|
||||
"organization": "Meta",
|
||||
"parameters": 70,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 8192,
|
||||
"input_price": 0,
|
||||
"output_price": 0,
|
||||
"mmlu": 82.0,
|
||||
"humaneval": 81.7,
|
||||
"is_open_source": true,
|
||||
"license": "Llama 3 Community",
|
||||
"description": "Meta最新开源模型，性能接近GPT-4",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "code"
|
||||
},
|
||||
{
|
||||
"id": "mistral7b",
|
||||
"name": "Mistral 7B",
|
||||
"organization": "Mistral AI",
|
||||
"parameters": 7,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 32768,
|
||||
"input_price": 0,
|
||||
"output_price": 0,
|
||||
"mmlu": 62.5,
|
||||
"humaneval": 26.8,
|
||||
"is_open_source": true,
|
||||
"license": "Apache 2.0",
|
||||
"description": "小巧高效的开源模型",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "mixtral8x7b",
|
||||
"name": "Mixtral 8x7B",
|
||||
"organization": "Mistral AI",
|
||||
"parameters": 47,
|
||||
"architecture": "MoE",
|
||||
"context_length": 32768,
|
||||
"input_price": 0,
|
||||
"output_price": 0,
|
||||
"mmlu": 70.6,
|
||||
"humaneval": 40.2,
|
||||
"is_open_source": true,
|
||||
"license": "Apache 2.0",
|
||||
"description": "MoE架构，高效推理",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "qwen72b",
|
||||
"name": "Qwen 72B",
|
||||
"organization": "Alibaba",
|
||||
"parameters": 72,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 32768,
|
||||
"input_price": 0,
|
||||
"output_price": 0,
|
||||
"mmlu": 83.1,
|
||||
"humaneval": 65.4,
|
||||
"is_open_source": true,
|
||||
"license": "Apache 2.0",
|
||||
"description": "阿里开源大模型，中文能力强",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "chat"
|
||||
},
|
||||
{
|
||||
"id": "deepseekv3",
|
||||
"name": "DeepSeek V3",
|
||||
"organization": "DeepSeek",
|
||||
"parameters": 685,
|
||||
"architecture": "MoE",
|
||||
"context_length": 128000,
|
||||
"input_price": 0.00014,
|
||||
"output_price": 0.00028,
|
||||
"mmlu": 88.5,
|
||||
"humaneval": 86.2,
|
||||
"is_open_source": true,
|
||||
"license": "MIT",
|
||||
"description": "DeepSeek最新模型，性价比极高",
|
||||
"created_at": "2024-01-01",
|
||||
"subcategory_id": "code"
|
||||
},
|
||||
{
|
||||
"id": "glm4",
|
||||
"name": "GLM-4",
|
||||
"organization": "Zhipu AI",
|
||||
"parameters": 130,
|
||||
"architecture": "Transformer",
|
||||
"context_length": 128000,
|
||||
"input_price": 0.014,
|
||||
"output_price": 0.014,
|
||||
"mmlu": 81.0,
|
||||
"humaneval": 70.0,
|
||||
"is_open_source": false,
|
||||
"license": "Proprietary",
|
||||
"description": "智谱AI大模型，中文能力强",
|
||||
"created_at": "2024-01-01",
|
||||
"visible": true,
|
||||
"subcategory_id": "chat"
|
||||
}
|
||||
]
|
||||
28303
logs/app.log
Normal file
28303
logs/app.log
Normal file
File diff suppressed because it is too large
Load Diff
14
static/favicon.svg
Normal file
14
static/favicon.svg
Normal file
@@ -0,0 +1,14 @@
|
||||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 32 32">
|
||||
<!-- 背景 -->
|
||||
<rect x="2" y="2" width="28" height="28" rx="6" fill="#4f46e5"/>
|
||||
<!-- 数据网格 -->
|
||||
<rect x="6" y="6" width="8" height="8" rx="2" fill="white"/>
|
||||
<rect x="18" y="6" width="8" height="8" rx="2" fill="white"/>
|
||||
<rect x="6" y="18" width="8" height="8" rx="2" fill="white"/>
|
||||
<rect x="18" y="18" width="8" height="8" rx="2" fill="white"/>
|
||||
<!-- 连接线 -->
|
||||
<line x1="14" y1="10" x2="18" y2="10" stroke="#4f46e5" stroke-width="2"/>
|
||||
<line x1="10" y1="14" x2="10" y2="18" stroke="#4f46e5" stroke-width="2"/>
|
||||
<line x1="22" y1="14" x2="22" y2="18" stroke="#4f46e5" stroke-width="2"/>
|
||||
<line x1="14" y1="22" x2="18" y2="22" stroke="#4f46e5" stroke-width="2"/>
|
||||
</svg>
|
||||
|
After Width: | Height: | Size: 757 B |
BIN
static/uploads/1ad784e0b3c6_1777305525.png
Normal file
BIN
static/uploads/1ad784e0b3c6_1777305525.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 1.1 MiB |
BIN
static/uploads/76f233ccdb91_1777287579.png
Normal file
BIN
static/uploads/76f233ccdb91_1777287579.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 1.1 MiB |
BIN
static/uploads/8db761aed139_1777286554.png
Normal file
BIN
static/uploads/8db761aed139_1777286554.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 1.1 MiB |
BIN
static/uploads/c6d7bedba2b0_1777286375.png
Normal file
BIN
static/uploads/c6d7bedba2b0_1777286375.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 289 B |
BIN
static/uploads/d2149794b5d3_1777286192.png
Normal file
BIN
static/uploads/d2149794b5d3_1777286192.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 289 B |
1893
templates/admin.html
1893
templates/admin.html
File diff suppressed because it is too large
Load Diff
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>{{ category.name }} - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -50,9 +51,16 @@
|
||||
class="w-full pl-12 pr-4 py-2 border border-gray-200 rounded-lg focus:outline-none focus:border-indigo-400"
|
||||
onkeyup="filterItems()">
|
||||
</div>
|
||||
<select id="sortSelect" onchange="sortItems()" class="px-4 py-2 border border-gray-200 rounded-lg focus:outline-none">
|
||||
<select id="sortBy" onchange="loadItems()" class="px-4 py-2 border border-gray-200 rounded-lg focus:outline-none">
|
||||
<option value="default">默认排序（置顶优先）</option>
|
||||
<option value="publish_date">按发布日期</option>
|
||||
<option value="views">按热度</option>
|
||||
<option value="name">按名称</option>
|
||||
<option value="created_at">按时间</option>
|
||||
<option value="created_at">按创建时间</option>
|
||||
</select>
|
||||
<select id="sortOrder" onchange="loadItems()" class="px-4 py-2 border border-gray-200 rounded-lg focus:outline-none">
|
||||
<option value="desc">降序</option>
|
||||
<option value="asc">升序</option>
|
||||
</select>
|
||||
</div>
|
||||
</div>
|
||||
@@ -115,7 +123,9 @@
|
||||
|
||||
// 加载数据
|
||||
async function loadItems() {
|
||||
const res = await fetch(`/api/items/${categoryId}`);
|
||||
const sortBy = document.getElementById('sortBy').value;
|
||||
const sortOrder = document.getElementById('sortOrder').value;
|
||||
const res = await fetch(`/api/items/${categoryId}?sort=${sortBy}&order=${sortOrder}`);
|
||||
allItems = await res.json();
|
||||
|
||||
document.getElementById('itemCount').textContent = allItems.length;
|
||||
@@ -136,22 +146,28 @@
|
||||
|
||||
document.getElementById('itemsList').innerHTML = items.map(item => {
|
||||
const fields = Object.entries(item)
|
||||
.filter(([key, val]) => !['id', 'category_id', 'created_at', 'updated_at'].includes(key) && val)
|
||||
.filter(([key, val]) => !['id', 'category_id', 'created_at', 'updated_at', 'visible', 'is_pinned', 'views', 'publish_date'].includes(key) && val)
|
||||
.slice(0, 5)
|
||||
.map(([key, val]) => `<span class="text-gray-500 text-sm">${key}: ${val}</span>`)
|
||||
.join('<br>');
|
||||
|
||||
return `
|
||||
<div class="border border-gray-200 rounded-lg p-4 hover:shadow-md transition group">
|
||||
<div class="border border-gray-200 rounded-lg p-4 hover:shadow-md transition group ${item.is_pinned ? 'bg-yellow-50 border-yellow-300' : ''}">
|
||||
<div class="flex items-start justify-between">
|
||||
<div>
|
||||
<h3 class="font-medium text-gray-800 group-hover:text-indigo-600">${item.name || item.title || '未命名'}</h3>
|
||||
<h3 class="font-medium text-gray-800 group-hover:text-indigo-600 flex items-center gap-2">
|
||||
${item.is_pinned ? '<i class="ri-pushpin-fill text-yellow-500" title="置顶"></i>' : ''}
|
||||
${item.name || item.title || '未命名'}
|
||||
</h3>
|
||||
<div class="mt-2 space-y-1">
|
||||
${fields}
|
||||
</div>
|
||||
</div>
|
||||
<div class="text-xs text-gray-400">
|
||||
${item.created_at ? item.created_at.split(' ')[0] : ''}
|
||||
<div class="text-right">
|
||||
<div class="text-xs text-gray-400">
|
||||
${item.publish_date || (item.created_at ? item.created_at.split(' ')[0] : '')}
|
||||
</div>
|
||||
${item.views ? `<div class="text-xs text-gray-400 mt-1"><i class="ri-eye-line"></i> ${item.views}</div>` : ''}
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
@@ -175,19 +191,6 @@
|
||||
renderItems(filtered);
|
||||
}
|
||||
|
||||
// 排序
|
||||
function sortItems() {
|
||||
const sortBy = document.getElementById('sortSelect').value;
|
||||
const sorted = [...allItems].sort((a, b) => {
|
||||
if (sortBy === 'name') {
|
||||
return (a.name || a.title || '').localeCompare(b.name || b.title || '');
|
||||
} else {
|
||||
return (b.created_at || '').localeCompare(a.created_at || '');
|
||||
}
|
||||
});
|
||||
renderItems(sorted);
|
||||
}
|
||||
|
||||
// 初始化
|
||||
loadNav();
|
||||
loadItems();
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>对比工具 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -40,7 +41,7 @@
|
||||
<i class="ri-cpu-line mr-2"></i>GPU对比
|
||||
</button>
|
||||
<button onclick="setCompareType('cpu')" id="btnCpu" class="px-4 py-2 bg-gray-200 text-gray-600 rounded-lg hover:bg-gray-300">
|
||||
<i class="ri-memory-line mr-2"></i>CPU对比
|
||||
<i class="ri-cpu-line mr-2"></i>CPU对比
|
||||
</button>
|
||||
</div>
|
||||
</div>
|
||||
@@ -230,6 +231,7 @@ let compareType = 'model';
|
||||
}
|
||||
|
||||
// 初始化
|
||||
loadNav();
|
||||
setCompareType('model');
|
||||
</script>
|
||||
</body>
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>CPU数据库 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -24,7 +25,7 @@
|
||||
<main class="max-w-7xl mx-auto px-4 py-8">
|
||||
<div class="mb-6">
|
||||
<h1 class="text-2xl font-bold text-gray-800 flex items-center gap-2">
|
||||
<i class="ri-memory-line text-purple-600"></i>
|
||||
<i class="ri-cpu-line text-purple-600"></i>
|
||||
CPU数据库
|
||||
</h1>
|
||||
<p class="text-gray-500 mt-1">处理器规格参数一览</p>
|
||||
@@ -208,6 +209,8 @@ async function loadCpus() {
|
||||
if (e.target === this) closeModal();
|
||||
});
|
||||
|
||||
// 初始化
|
||||
loadNav();
|
||||
loadCpus();
|
||||
</script>
|
||||
</body>
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>GPU数据库 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -215,6 +216,8 @@ async function loadGpus() {
|
||||
if (e.target === this) closeModal();
|
||||
});
|
||||
|
||||
// 初始化
|
||||
loadNav();
|
||||
loadGpus();
|
||||
</script>
|
||||
</body>
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<title id="pageTitle">ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -13,8 +14,8 @@
|
||||
<div class="max-w-7xl mx-auto px-4 py-3 flex justify-between items-center">
|
||||
<div class="flex items-center gap-2">
|
||||
<i class="ri-dashboard-3-line text-2xl text-indigo-600"></i>
|
||||
<span class="text-xl font-bold text-gray-800">ParamHub</span>
|
||||
<span class="text-sm text-gray-500">参数百科</span>
|
||||
<span class="text-xl font-bold text-gray-800" id="navSiteName">ParamHub</span>
|
||||
<span class="text-sm text-gray-500" id="navSiteSubtitle">参数百科</span>
|
||||
</div>
|
||||
<div class="flex gap-4 text-sm" id="navLinks">
|
||||
<!-- 动态加载 -->
|
||||
@@ -58,7 +59,7 @@
|
||||
<!-- 热门模型 -->
|
||||
<div class="bg-white rounded-xl shadow-sm p-6">
|
||||
<h2 class="text-lg font-semibold text-gray-800 mb-4 flex items-center gap-2">
|
||||
<i class="ri-flashlight-line text-indigo-600"></i>
|
||||
<i class="ri-cpu-line text-indigo-600"></i>
|
||||
热门模型
|
||||
</h2>
|
||||
<div id="latestModels" class="grid grid-cols-2 gap-4">
|
||||
@@ -69,7 +70,7 @@
|
||||
|
||||
<!-- 页脚 -->
|
||||
<footer class="bg-white border-t mt-8 py-6 text-center text-gray-500 text-sm">
|
||||
ParamHub - 参数百科 | AI模型与硬件参数速查平台
|
||||
<span id="footerText">ParamHub - 参数百科 | AI模型与硬件参数速查平台</span>
|
||||
</footer>
|
||||
|
||||
<script>
|
||||
@@ -120,13 +121,32 @@
|
||||
|
||||
// 加载分类和数据统计
|
||||
async function loadData() {
|
||||
// 并行加载分类和统计
|
||||
const [catRes, statsRes] = await Promise.all([
|
||||
// 并行加载分类、统计、配置
|
||||
const [catRes, statsRes, configRes] = await Promise.all([
|
||||
fetch('/api/categories'),
|
||||
fetch('/api/stats')
|
||||
fetch('/api/stats'),
|
||||
fetch('/api/config')
|
||||
]);
|
||||
categories = await catRes.json();
|
||||
const stats = await statsRes.json();
|
||||
const config = await configRes.json();
|
||||
|
||||
// 更新网站名称等
|
||||
if (config.site_name) {
|
||||
document.getElementById('navSiteName').textContent = config.site_name;
|
||||
}
|
||||
if (config.site_subtitle) {
|
||||
document.getElementById('navSiteSubtitle').textContent = config.site_subtitle;
|
||||
}
|
||||
if (config.site_name && config.site_subtitle) {
|
||||
document.getElementById('pageTitle').textContent = `${config.site_name} - ${config.site_subtitle}`;
|
||||
}
|
||||
if (config.footer_text) {
|
||||
document.getElementById('footerText').textContent = config.footer_text;
|
||||
}
|
||||
if (config.icp_number) {
|
||||
document.getElementById('footerText').innerHTML += ` | <a href="https://beian.miit.gov.cn/" target="_blank" class="hover:text-gray-700">${config.icp_number}</a>`;
|
||||
}
|
||||
|
||||
// 渲染导航栏
|
||||
renderNavBar();
|
||||
@@ -146,7 +166,7 @@
|
||||
const statItems = [
|
||||
{key: 'models_count', label: 'AI模型', icon: 'ri-robot-line', color: 'blue'},
|
||||
{key: 'gpus_count', label: 'GPU显卡', icon: 'ri-cpu-line', color: 'green'},
|
||||
{key: 'cpus_count', label: 'CPU处理器', icon: 'ri-memory-line', color: 'purple'},
|
||||
{key: 'cpus_count', label: 'CPU处理器', icon: 'ri-cpu-line', color: 'purple'},
|
||||
{key: 'knowledge_count', label: '知识条目', icon: 'ri-book-open-line', color: 'teal'}
|
||||
];
|
||||
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>知识库 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -75,7 +76,7 @@
|
||||
<!-- 显存计算 -->
|
||||
<div class="bg-white rounded-xl shadow-sm p-6">
|
||||
<h2 class="text-lg font-semibold text-gray-800 mb-4 flex items-center gap-2">
|
||||
<i class="ri-memory-line text-orange-600"></i>
|
||||
<i class="ri-cpu-line text-orange-600"></i>
|
||||
如何计算显存需求？
|
||||
</h2>
|
||||
<p class="text-gray-600 leading-relaxed">
|
||||
@@ -177,5 +178,43 @@
|
||||
</table>
|
||||
</div>
|
||||
</main>
|
||||
|
||||
<script>
|
||||
let categories = [];
|
||||
|
||||
// 加载导航栏
|
||||
async function loadNav() {
|
||||
const res = await fetch('/api/categories');
|
||||
categories = await res.json();
|
||||
|
||||
const builtinPages = [
|
||||
{name: '首页', href: '/'},
|
||||
{name: '工具', href: '/tools'},
|
||||
{name: '对比', href: '/compare'},
|
||||
{name: '知识库', href: '/knowledge'}
|
||||
];
|
||||
|
||||
const categoryPages = {
|
||||
'ai-models': '/models',
|
||||
'gpus': '/gpus',
|
||||
'cpus': '/cpus'
|
||||
};
|
||||
|
||||
let navHtml = '';
|
||||
builtinPages.forEach(p => {
|
||||
const isActive = window.location.pathname === p.href;
|
||||
navHtml += `<a href="${p.href}" class="${isActive ? 'text-indigo-600 font-medium' : 'text-gray-600 hover:text-indigo-600'}">${p.name}</a>`;
|
||||
});
|
||||
|
||||
categories.forEach(cat => {
|
||||
const href = categoryPages[cat.id] || `/category/${cat.id}`;
|
||||
navHtml += `<a href="${href}" class="text-gray-600 hover:text-indigo-600">${cat.name}</a>`;
|
||||
});
|
||||
|
||||
document.getElementById('topNav').innerHTML = navHtml;
|
||||
}
|
||||
|
||||
loadNav();
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>模型数据库 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -31,6 +32,19 @@
|
||||
<p class="text-gray-500 mt-1">AI大模型参数规格一览</p>
|
||||
</div>
|
||||
|
||||
<!-- 子类别选择器 -->
|
||||
<div class="bg-white rounded-xl shadow-sm p-4 mb-4">
|
||||
<div class="flex items-center gap-2 mb-2">
|
||||
<span class="text-sm text-gray-600"><i class="ri-folder-line mr-1"></i>子类别：</span>
|
||||
</div>
|
||||
<div class="flex gap-2" id="subcategoryTabs">
|
||||
<button onclick="selectSubcategory('')" class="px-4 py-2 bg-indigo-600 text-white rounded-lg text-sm" id="subcat-all">
|
||||
<i class="ri-apps-line mr-1"></i>全部
|
||||
</button>
|
||||
<!-- 动态加载子类别 -->
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<!-- 搜索和筛选 -->
|
||||
<div class="bg-white rounded-xl shadow-sm p-4 mb-6">
|
||||
<div class="flex gap-4 items-center">
|
||||
@@ -41,14 +55,18 @@
|
||||
oninput="loadModels()">
|
||||
</div>
|
||||
<select id="sortBy" class="px-4 py-2 border border-gray-200 rounded-lg" onchange="loadModels()">
|
||||
<option value="default">默认排序（置顶优先）</option>
|
||||
<option value="publish_date">按发布日期</option>
|
||||
<option value="views">按热度</option>
|
||||
<option value="name">按名称</option>
|
||||
<option value="parameters">按参数量</option>
|
||||
<option value="mmlu">按MMLU分数</option>
|
||||
<option value="context_length">按上下文长度</option>
|
||||
<option value="created_at">按创建时间</option>
|
||||
</select>
|
||||
<select id="sortOrder" class="px-4 py-2 border border-gray-200 rounded-lg" onchange="loadModels()">
|
||||
<option value="asc">升序</option>
|
||||
<option value="desc">降序</option>
|
||||
<option value="asc">升序</option>
|
||||
</select>
|
||||
<select id="filterType" class="px-4 py-2 border border-gray-200 rounded-lg" onchange="loadModels()">
|
||||
<option value="all">全部</option>
|
||||
@@ -96,12 +114,38 @@
|
||||
<script>
|
||||
let allModels = [];
|
||||
let categories = [];
|
||||
let currentCategory = null;
|
||||
let currentSubcategory = '';
|
||||
|
||||
// 子类别默认特性配置
|
||||
const DEFAULT_KEY_FEATURES = {
|
||||
'chat': ['context_length', 'mmlu', 'input_price', 'output_price'],
|
||||
'code': ['humaneval', 'context_length', 'input_price'],
|
||||
'reasoning': ['mmlu', 'context_length', 'parameters'],
|
||||
'vision': ['context_length', 'mmlu', 'input_price']
|
||||
};
|
||||
|
||||
const FEATURE_LABELS = {
|
||||
'context_length': '上下文',
|
||||
'mmlu': 'MMLU',
|
||||
'humaneval': 'HumanEval',
|
||||
'input_price': '输入价',
|
||||
'output_price': '输出价',
|
||||
'parameters': '参数量',
|
||||
'reasoning_capability': '推理',
|
||||
'vision_capability': '视觉',
|
||||
'multimodal': '多模态'
|
||||
};
|
||||
|
||||
// 加载导航栏
|
||||
async function loadNav() {
|
||||
const res = await fetch('/api/categories');
|
||||
categories = await res.json();
|
||||
|
||||
// 获取当前类别的子类别
|
||||
currentCategory = categories.find(c => c.id === 'ai-models');
|
||||
renderSubcategoryTabs();
|
||||
|
||||
const builtinPages = [
|
||||
{name: '首页', href: '/'},
|
||||
{name: '工具', href: '/tools'},
|
||||
@@ -129,6 +173,49 @@
|
||||
|
||||
document.getElementById('topNav').innerHTML = navHtml;
|
||||
}
|
||||
|
||||
// 渲染子类别选择器
|
||||
function renderSubcategoryTabs() {
|
||||
const container = document.getElementById('subcategoryTabs');
|
||||
if (!currentCategory || !currentCategory.subcategories) {
|
||||
container.innerHTML = '';
|
||||
return;
|
||||
}
|
||||
|
||||
let html = `<button onclick="selectSubcategory('')" class="px-4 py-2 ${currentSubcategory === '' ? 'bg-indigo-600 text-white' : 'bg-gray-100 text-gray-600 hover:bg-gray-200'} rounded-lg text-sm" id="subcat-all">
|
||||
<i class="ri-apps-line mr-1"></i>全部
|
||||
</button>`;
|
||||
|
||||
currentCategory.subcategories.forEach(sub => {
|
||||
const isActive = currentSubcategory === sub.id;
|
||||
html += `<button onclick="selectSubcategory('${sub.id}')" class="px-4 py-2 ${isActive ? 'bg-indigo-600 text-white' : 'bg-gray-100 text-gray-600 hover:bg-gray-200'} rounded-lg text-sm" id="subcat-${sub.id}">
|
||||
<i class="${sub.icon || 'ri-folder-line'} mr-1"></i>${sub.name}
|
||||
</button>`;
|
||||
});
|
||||
|
||||
container.innerHTML = html;
|
||||
}
|
||||
|
||||
// 选择子类别
|
||||
function selectSubcategory(subId) {
|
||||
currentSubcategory = subId;
|
||||
renderSubcategoryTabs();
|
||||
loadModels();
|
||||
}
|
||||
|
||||
// 获取当前子类别的关键特性
|
||||
function getKeyFeatures() {
|
||||
if (!currentSubcategory || !currentCategory || !currentCategory.subcategories) {
|
||||
return ['parameters', 'context_length', 'mmlu', 'input_price'];
|
||||
}
|
||||
|
||||
const subcat = currentCategory.subcategories.find(s => s.id === currentSubcategory);
|
||||
if (subcat && subcat.key_features) {
|
||||
return subcat.key_features;
|
||||
}
|
||||
|
||||
return ['parameters', 'context_length', 'mmlu', 'input_price'];
|
||||
}
|
||||
|
||||
async function loadModels() {
|
||||
const keyword = document.getElementById('searchInput').value.trim();
|
||||
@@ -150,6 +237,21 @@
|
||||
models = models.filter(m => !m.is_open_source);
|
||||
}
|
||||
|
||||
// 子类别过滤（通过模型名称/描述中的关键词判断）
|
||||
if (currentSubcategory && currentCategory && currentCategory.subcategories) {
|
||||
const subcat = currentCategory.subcategories.find(s => s.id === currentSubcategory);
|
||||
if (subcat) {
|
||||
// 简化过滤：根据子类别关键词匹配
|
||||
// 实际应该有 subcategory_id 字段，这里暂时用名称匹配
|
||||
// 用户可以在后台编辑时指定子类别
|
||||
models = models.filter(m => {
|
||||
const subcatField = m.subcategory || m.subcategory_id;
|
||||
if (subcatField) return subcatField === currentSubcategory;
|
||||
return true; // 暂时显示全部，等后台支持子类别字段后再过滤
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
renderModels(models);
|
||||
}
|
||||
|
||||
@@ -161,38 +263,93 @@
|
||||
return;
|
||||
}
|
||||
|
||||
const html = models.map(m => `
|
||||
<tr class="border-b hover:bg-gray-50 transition">
|
||||
// 动态获取关键特性
|
||||
const keyFeatures = getKeyFeatures();
|
||||
|
||||
// 动态表头
|
||||
let headerHtml = `
|
||||
<tr>
|
||||
<th class="px-4 py-3 text-left text-sm font-medium text-gray-600">模型名称</th>
|
||||
<th class="px-4 py-3 text-left text-sm font-medium text-gray-600">厂商</th>
|
||||
`;
|
||||
|
||||
keyFeatures.forEach(f => {
|
||||
headerHtml += `<th class="px-4 py-3 text-left text-sm font-medium text-gray-600">${FEATURE_LABELS[f] || f}</th>`;
|
||||
});
|
||||
|
||||
headerHtml += `
|
||||
<th class="px-4 py-3 text-left text-sm font-medium text-gray-600">类型</th>
|
||||
<th class="px-4 py-3 text-center text-sm font-medium text-gray-600">操作</th>
|
||||
</tr>
|
||||
`;
|
||||
|
||||
document.querySelector('#modelsTable thead').innerHTML = headerHtml;
|
||||
|
||||
// 动态内容
|
||||
const html = models.map(m => {
|
||||
let rowHtml = `
|
||||
<tr class="border-b hover:bg-gray-50 transition ${m.is_pinned ? 'bg-yellow-50' : ''}">
|
||||
<td class="px-4 py-3">
|
||||
<div class="font-medium text-gray-800">${m.name}</div>
|
||||
<div class="text-xs text-gray-500">${m.architecture || ''}</div>
|
||||
<div class="flex items-center gap-2">
|
||||
${m.is_pinned ? '<i class="ri-pushpin-fill text-yellow-500" title="置顶"></i>' : ''}
|
||||
<div>
|
||||
<div class="font-medium text-gray-800">${m.name}</div>
|
||||
<div class="text-xs text-gray-500">${m.architecture || ''}</div>
|
||||
</div>
|
||||
</div>
|
||||
</td>
|
||||
<td class="px-4 py-3 text-gray-600">${m.organization}</td>
|
||||
<td class="px-4 py-3">
|
||||
<span class="px-2 py-1 bg-blue-100 text-blue-700 rounded text-sm">${m.parameters}B</span>
|
||||
</td>
|
||||
<td class="px-4 py-3 text-gray-600">${formatContext(m.context_length)}</td>
|
||||
<td class="px-4 py-3">
|
||||
<span class="px-2 py-1 bg-green-100 text-green-700 rounded text-sm">${m.mmlu || '-'}%</span>
|
||||
</td>
|
||||
`;
|
||||
|
||||
// 关键特性列
|
||||
keyFeatures.forEach(f => {
|
||||
const value = formatFeatureValue(f, m);
|
||||
rowHtml += `<td class="px-4 py-3">${value}</td>`;
|
||||
});
|
||||
|
||||
rowHtml += `
|
||||
<td class="px-4 py-3">
|
||||
${m.is_open_source
|
||||
? '<span class="px-2 py-1 bg-emerald-100 text-emerald-700 rounded text-sm">开源</span>'
|
||||
: '<span class="px-2 py-1 bg-gray-100 text-gray-700 rounded text-sm">商业</span>'}
|
||||
</td>
|
||||
<td class="px-4 py-3 text-sm text-gray-600">
|
||||
${m.input_price ? `$${m.input_price}/$${m.output_price}` : '免费'}
|
||||
</td>
|
||||
<td class="px-4 py-3 text-center">
|
||||
<button onclick="showDetail('${m.id}')" class="text-indigo-600 hover:text-indigo-800 text-sm">
|
||||
<i class="ri-eye-line mr-1"></i>详情
|
||||
</button>
|
||||
</td>
|
||||
</tr>
|
||||
`).join('');
|
||||
`;
|
||||
|
||||
return rowHtml;
|
||||
}).join('');
|
||||
|
||||
document.getElementById('modelsTable').innerHTML = html;
|
||||
}
|
||||
|
||||
// 格式化特性值
|
||||
function formatFeatureValue(feature, model) {
|
||||
const value = model[feature];
|
||||
|
||||
if (value === null || value === undefined) return '<span class="text-gray-400">-</span>';
|
||||
|
||||
switch (feature) {
|
||||
case 'parameters':
|
||||
return `<span class="px-2 py-1 bg-blue-100 text-blue-700 rounded text-sm">${value}B</span>`;
|
||||
case 'context_length':
|
||||
return `<span class="text-gray-600">${formatContext(value)}</span>`;
|
||||
case 'mmlu':
|
||||
return `<span class="px-2 py-1 bg-green-100 text-green-700 rounded text-sm">${value}%</span>`;
|
||||
case 'humaneval':
|
||||
return `<span class="px-2 py-1 bg-purple-100 text-purple-700 rounded text-sm">${value}%</span>`;
|
||||
case 'input_price':
|
||||
return `<span class="text-sm text-gray-600">$${value || 0}</span>`;
|
||||
case 'output_price':
|
||||
return `<span class="text-sm text-gray-600">$${value || 0}</span>`;
|
||||
default:
|
||||
return `<span class="text-gray-600">${value}</span>`;
|
||||
}
|
||||
}
|
||||
|
||||
function formatContext(len) {
|
||||
if (!len) return '-';
|
||||
|
||||
@@ -3,7 +3,8 @@
|
||||
<head>
|
||||
<meta charset="UTF-8">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<title>实用工具 - ParamHub</title>
|
||||
<title>ParamHub - 参数百科</title>
|
||||
<link rel="icon" type="image/svg+xml" href="/static/favicon.svg">
|
||||
<script src="https://cdn.tailwindcss.com"></script>
|
||||
<link href="https://cdn.jsdelivr.net/npm/remixicon@3.5.0/fonts/remixicon.css" rel="stylesheet">
|
||||
</head>
|
||||
@@ -33,7 +34,7 @@
|
||||
<!-- 显存计算器 -->
|
||||
<div class="bg-white rounded-xl shadow-sm p-6 mb-6">
|
||||
<h2 class="text-lg font-semibold text-gray-800 mb-4 flex items-center gap-2">
|
||||
<i class="ri-memory-line text-green-600"></i>
|
||||
<i class="ri-cpu-line text-green-600"></i>
|
||||
显存计算器
|
||||
</h2>
|
||||
<p class="text-gray-500 mb-4">计算大模型所需的显存大小，并推荐合适的GPU</p>
|
||||
@@ -197,6 +198,9 @@ async function calculateVram() {
|
||||
|
||||
document.getElementById('vramResult').classList.remove('hidden');
|
||||
}
|
||||
|
||||
// 初始化
|
||||
loadNav();
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
Reference in New Issue
Block a user