{
"schemaVersion": 2,
"generatedAt": "2026-09-01T02:44:54.336Z",
"models": [
{
"id": "gemma4:e4b-it-qat",
"family": "gemma4",
"params": "8B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"vision",
"code",
"tools",
"reasoning"
],
"releaseDate": "2026-03-31",
"source": {
"ollama": "gemma4:e4b-it-qat",
"hf": "google/gemma-4-E4B-it-qat-q4_0-gguf"
},
"quantizations": [
{
"name": "Q4_0",
"diskBytes": 5154939136,
"minRamBytes": 5928180007,
"minVramBytes": 5928180007,
"sha256": "e8b6a059ba86947a44ace84d6e5679795bc41862c25c30513142588f0e9dba1d"
}
],
"benchmarkProxy": 0.52
},
{
"id": "kimi-k2-thinking",
"family": "kimi-k2-thinking",
"params": "1T",
"architecture": "moe",
"activeParams": "32B",
"license": "modified-mit",
"openWeight": true,
"contextLength": 262144,
"capabilities": [
"chat",
"reasoning",
"tools"
],
"releaseDate": "2025-11-06",
"source": {
"hf": "moonshotai/Kimi-K2-Thinking"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 600000000000,
"minRamBytes": 690000000000,
"minVramBytes": 690000000000
}
],
"benchmarkProxy": 0.99
},
{
"id": "kimi-linear",
"family": "kimi-linear",
"params": "48B",
"architecture": "moe",
"activeParams": "3B",
"license": "modified-mit",
"openWeight": true,
"contextLength": 262144,
"capabilities": [
"chat",
"reasoning"
],
"releaseDate": "2025-10-30",
"source": {
"hf": "moonshotai/Kimi-Linear-48B-A3B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 29000000000,
"minRamBytes": 33350000000,
"minVramBytes": 33350000000
}
],
"benchmarkProxy": 0.78
},
{
"id": "kimi-k2:base",
"family": "kimi-k2",
"params": "1T",
"architecture": "moe",
"activeParams": "32B",
"license": "modified-mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2025-07-11",
"source": {
"hf": "moonshotai/Kimi-K2-Base"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 600000000000,
"minRamBytes": 690000000000,
"minVramBytes": 690000000000
}
],
"benchmarkProxy": 0.97
},
{
"id": "kimi-k2:instruct",
"family": "kimi-k2",
"params": "1T",
"architecture": "moe",
"activeParams": "32B",
"license": "modified-mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-07-11",
"source": {
"ollama": "kimi-k2:instruct",
"hf": "moonshotai/Kimi-K2-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 600000000000,
"minRamBytes": 690000000000,
"minVramBytes": 690000000000
}
],
"benchmarkProxy": 0.97
},
{
"id": "gemma3n:e2b",
"family": "gemma3n",
"params": "6B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"vision",
"code"
],
"releaseDate": "2025-06-26",
"source": {
"ollama": "gemma3n:e2b",
"hf": "google/gemma-3n-E2B-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5621607424,
"minRamBytes": 6464848538,
"minVramBytes": 6464848538,
"sha256": "3839a254cf2d00b208c6e2524c129e4438f9d106bba4c3fbc12b631f519d1de1"
}
],
"benchmarkProxy": 0.42
},
{
"id": "gemma3n:e4b",
"family": "gemma3n",
"params": "8B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"vision",
"code"
],
"releaseDate": "2025-06-26",
"source": {
"ollama": "gemma3n:e4b",
"hf": "google/gemma-3n-E4B-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 7547579904,
"minRamBytes": 8679716890,
"minVramBytes": 8679716890,
"sha256": "38e8dcc30df4eb0e29eaf5c74ba6ce3f2cd66badad50768fc14362acfb8b8cb6"
}
],
"benchmarkProxy": 0.5
},
{
"id": "kimi-dev-72b",
"family": "kimi-dev",
"params": "72B",
"architecture": "dense",
"license": "modified-mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"code",
"chat",
"reasoning"
],
"releaseDate": "2025-06-16",
"source": {
"hf": "moonshotai/Kimi-Dev-72B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 43000000000,
"minRamBytes": 49450000000,
"minVramBytes": 49450000000
}
],
"benchmarkProxy": 0.78
},
{
"id": "qwen3:0.6b",
"family": "qwen3",
"params": "0.6B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 40960,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-29",
"source": {
"ollama": "qwen3:0.6b",
"hf": "Qwen/Qwen3-0.6B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 522640096,
"minRamBytes": 601036111,
"minVramBytes": 601036111,
"sha256": "7f4030143c1c477224c5434f8272c662a8b042079a0a584f0a27a1684fe2e1fa"
}
],
"benchmarkProxy": 0.22
},
{
"id": "qwen3:1.7b",
"family": "qwen3",
"params": "1.7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 40960,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-29",
"source": {
"ollama": "qwen3:1.7b",
"hf": "Qwen/Qwen3-1.7B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1359279776,
"minRamBytes": 1563171743,
"minVramBytes": 1563171743,
"sha256": "3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6"
}
],
"benchmarkProxy": 0.3
},
{
"id": "qwen3:4b",
"family": "qwen3",
"params": "4B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 262144,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-29",
"source": {
"ollama": "qwen3:4b",
"hf": "Qwen/Qwen3-4B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2497280480,
"minRamBytes": 2871872552,
"minVramBytes": 2871872552,
"sha256": "3e4cb14174460404e7a233e531675303b2fbf7749c02f91864fe311ab6344e4f"
}
],
"benchmarkProxy": 0.46
},
{
"id": "qwen3:14b",
"family": "qwen3",
"params": "14B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 40960,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-28",
"source": {
"ollama": "qwen3:14b",
"hf": "Qwen/Qwen3-14B",
"gguf": {
"repo": "Qwen/Qwen3-14B-GGUF",
"revision": "530227a7d994db8eca5ab5ced2fb692b614357fd",
"file": "Qwen3-14B-Q4_K_M.gguf",
"sha256": "500a8806e85ee9c83f3ae08420295592451379b4f8cf2d0f41c15dffeb6b81f0"
}
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 9276184896,
"minRamBytes": 10667612631,
"minVramBytes": 10667612631,
"sha256": "a8cc1361f3145dc01f6d77c6c82c9116b9ffe3c97b34716fe20418455876c40e"
}
],
"benchmarkProxy": 0.62
},
{
"id": "qwen3:235b-a22b",
"family": "qwen3",
"params": "235B",
"architecture": "moe",
"activeParams": "22B",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 262144,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-28",
"source": {
"ollama": "qwen3:235b-a22b",
"hf": "Qwen/Qwen3-235B-A22B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 142154073664,
"minRamBytes": 163477184714,
"minVramBytes": 163477184714,
"sha256": "791d5d11998e006548d6b58c31756562ea61446ebc7d19686608402a797ecc82"
}
],
"benchmarkProxy": 0.95
},
{
"id": "qwen3:30b-a3b",
"family": "qwen3",
"params": "30B",
"architecture": "moe",
"activeParams": "3B",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 262144,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-28",
"source": {
"ollama": "qwen3:30b-a3b",
"hf": "Qwen/Qwen3-30B-A3B",
"gguf": {
"repo": "Qwen/Qwen3-30B-A3B-GGUF",
"revision": "e4d4bafdfb96a411a163846265362aceb0b9c63a",
"file": "Qwen3-30B-A3B-Q4_K_M.gguf",
"sha256": "0d003f6662faee786ed5da3e31b29c978de5ae5d275c8794c606a7f3c01aa8f5"
}
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 18556685856,
"minRamBytes": 21340188735,
"minVramBytes": 21340188735,
"sha256": "58574f2e94b99fb9e4391408b57e5aeaaaec10f6384e9a699fc2cb43a5c8eabf"
}
],
"benchmarkProxy": 0.74
},
{
"id": "qwen3:32b",
"family": "qwen3",
"params": "32B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 40960,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-28",
"source": {
"ollama": "qwen3:32b",
"hf": "Qwen/Qwen3-32B",
"gguf": {
"repo": "Qwen/Qwen3-32B-GGUF",
"revision": "938a7432affaec9157f883a87164e2646ae17555",
"file": "Qwen3-32B-Q4_K_M.gguf",
"sha256": "efd971561896866f0e910cce52761ca77b1b138090c7f15fe284676d57d1f689"
}
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 20201240160,
"minRamBytes": 23231426184,
"minVramBytes": 23231426184,
"sha256": "3291abe70f16ee9682de7bfae08db5373ea9d6497e614aaad63340ad421d6312"
}
],
"benchmarkProxy": 0.74
},
{
"id": "qwen3:8b",
"family": "qwen3",
"params": "8B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 40960,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-04-28",
"source": {
"ollama": "qwen3:8b",
"hf": "Qwen/Qwen3-8B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5225374496,
"minRamBytes": 6009180671,
"minVramBytes": 6009180671,
"sha256": "a3de86cd1c132c822487ededd47a324c50491393e6565cd14bafa40d0b8e686f"
}
],
"benchmarkProxy": 0.54
},
{
"id": "glm4:9b",
"family": "glm4",
"params": "9B",
"architecture": "dense",
"license": "mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2025-04-14",
"source": {
"ollama": "glm4:9b",
"hf": "THUDM/GLM-4-9B-0414"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5455319040,
"minRamBytes": 6273616896,
"minVramBytes": 6273616896,
"sha256": "b506a070d1152798d435ec4e7687336567ae653b3106f73b7b4ac7be1cbc4449"
}
],
"benchmarkProxy": 0.49
},
{
"id": "kimi-vl-a3b",
"family": "kimi-vl",
"params": "16B",
"architecture": "moe",
"activeParams": "3B",
"license": "modified-mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"vision"
],
"releaseDate": "2025-04-10",
"source": {
"hf": "moonshotai/Kimi-VL-A3B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 10000000000,
"minRamBytes": 11500000000,
"minVramBytes": 11500000000
}
],
"benchmarkProxy": 0.64
},
{
"id": "gemma3:12b",
"family": "gemma3",
"params": "12B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"vision",
"code"
],
"releaseDate": "2025-03-12",
"source": {
"ollama": "gemma3:12b",
"hf": "google/gemma-3-12b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 8149180896,
"minRamBytes": 9371558031,
"minVramBytes": 9371558031,
"sha256": "e8ad13eff07a78d89926e9e8b882317d082ef5bf9768ad7b50fcdbbcd63748de"
}
],
"benchmarkProxy": 0.58
},
{
"id": "gemma3:27b",
"family": "gemma3",
"params": "27B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"vision",
"code",
"reasoning"
],
"releaseDate": "2025-03-12",
"source": {
"ollama": "gemma3:27b",
"hf": "google/gemma-3-27b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 17396927584,
"minRamBytes": 20006466722,
"minVramBytes": 20006466722,
"sha256": "e796792eba26c4d3b04b0ac5adb01a453dd9ec2dfd83b6c59cbf6fe5f30b0f68"
}
],
"benchmarkProxy": 0.7
},
{
"id": "gemma3:4b",
"family": "gemma3",
"params": "4B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"vision"
],
"releaseDate": "2025-03-12",
"source": {
"ollama": "gemma3:4b",
"hf": "google/gemma-3-4b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 3338792448,
"minRamBytes": 3839611316,
"minVramBytes": 3839611316,
"sha256": "aeda25e63ebd698fab8638ffb778e68bed908b960d39d0becc650fa981609d25"
}
],
"benchmarkProxy": 0.42
},
{
"id": "phi4-mini:3.8b",
"family": "phi4-mini",
"params": "3.8B",
"architecture": "dense",
"license": "mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"tools",
"code"
],
"releaseDate": "2025-02-26",
"source": {
"ollama": "phi4-mini:3.8b",
"hf": "microsoft/Phi-4-mini-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2491874624,
"minRamBytes": 2865655818,
"minVramBytes": 2865655818,
"sha256": "3c168af1dea0a414299c7d9077e100ac763370e5a98b3c53801a958a47f0a5db"
}
],
"benchmarkProxy": 0.38
},
{
"id": "mistral-small:24b",
"family": "mistral-small",
"params": "24B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2025-01-30",
"source": {
"ollama": "mistral-small:24b",
"hf": "mistralai/Mistral-Small-24B-Instruct-2501"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 14333908384,
"minRamBytes": 16483994642,
"minVramBytes": 16483994642,
"sha256": "102a747c137683e81d431dab05d8f2158df4ab6f162f8f9019425a43d51e0e9f"
}
],
"kvBytesPerToken": 163840,
"benchmarkProxy": 0.63
},
{
"id": "deepseek-r1:14b",
"family": "deepseek-r1",
"params": "14B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:14b",
"hf": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 8988109952,
"minRamBytes": 10336326445,
"minVramBytes": 10336326445,
"sha256": "6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e"
}
],
"benchmarkProxy": 0.64
},
{
"id": "deepseek-r1:32b",
"family": "deepseek-r1",
"params": "32B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:32b",
"hf": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 19851335552,
"minRamBytes": 22829035885,
"minVramBytes": 22829035885,
"sha256": "6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93"
}
],
"benchmarkProxy": 0.76
},
{
"id": "deepseek-r1:671b",
"family": "deepseek-r1",
"params": "671B",
"architecture": "moe",
"activeParams": "37B",
"license": "mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"reasoning",
"chat",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:671b",
"hf": "deepseek-ai/DeepSeek-R1"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 404430186848,
"minRamBytes": 465094714876,
"minVramBytes": 465094714876,
"sha256": "439dd1a5e05286918f54941f49f9b56118c757440f6333f67f1cd5cbb5c8520b"
}
],
"benchmarkProxy": 0.97
},
{
"id": "deepseek-r1:70b",
"family": "deepseek-r1",
"params": "70B",
"architecture": "dense",
"license": "llama-3.3-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:70b",
"hf": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 42520395616,
"minRamBytes": 48898454959,
"minVramBytes": 48898454959,
"sha256": "4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339"
}
],
"benchmarkProxy": 0.82
},
{
"id": "deepseek-r1:7b",
"family": "deepseek-r1",
"params": "7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:7b",
"hf": "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4683073184,
"minRamBytes": 5385534162,
"minVramBytes": 5385534162,
"sha256": "96c415656d377afbff962f6cdb2394ab092ccbcbaab4b82525bc4ca800fe8a49"
}
],
"benchmarkProxy": 0.56
},
{
"id": "deepseek-r1:8b",
"family": "deepseek-r1",
"params": "8B",
"architecture": "dense",
"license": "llama-3.1-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"reasoning",
"code"
],
"releaseDate": "2025-01-20",
"source": {
"ollama": "deepseek-r1:8b",
"hf": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5225373760,
"minRamBytes": 6009179824,
"minVramBytes": 6009179824,
"sha256": "e6a7edc1a4d7d9b2de136a221a57336b76316cfe53a252aeba814496c5ae439d"
}
],
"benchmarkProxy": 0.56
},
{
"id": "deepseek-v3",
"family": "deepseek-v3",
"params": "671B",
"architecture": "moe",
"activeParams": "37B",
"license": "deepseek",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-12-26",
"source": {
"ollama": "deepseek-v3",
"hf": "deepseek-ai/DeepSeek-V3"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 404430186016,
"minRamBytes": 465094713919,
"minVramBytes": 465094713919,
"sha256": "d83c18fb2a2ccca56d641920f01e6fe533dfb3fcf4b77c007c931497cd24a517"
}
],
"benchmarkProxy": 0.95
},
{
"id": "granite3.1:2b",
"family": "granite3.1",
"params": "2B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-12-18",
"source": {
"ollama": "granite3.1-dense:2b",
"hf": "ibm-granite/granite-3.1-2b-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1569676320,
"minRamBytes": 1805127768,
"minVramBytes": 1805127768,
"sha256": "5c56bb0256a2c402e95282a29bb5cb747bb805eda0e14a84b1f6c594a297ec1a"
}
],
"benchmarkProxy": 0.31
},
{
"id": "granite3.1:8b",
"family": "granite3.1",
"params": "8B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-12-18",
"source": {
"ollama": "granite3.1-dense:8b",
"hf": "ibm-granite/granite-3.1-8b-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4991620512,
"minRamBytes": 5740363589,
"minVramBytes": 5740363589,
"sha256": "44d19d212d76a6f3fc442e8411fdb44ea6b67ceccfb00be4b4345c9a4cf813e8"
}
],
"benchmarkProxy": 0.47
},
{
"id": "phi4:14b",
"family": "phi4",
"params": "14B",
"architecture": "dense",
"license": "mit",
"openWeight": true,
"contextLength": 16384,
"capabilities": [
"chat",
"code",
"reasoning"
],
"releaseDate": "2024-12-12",
"source": {
"ollama": "phi4:14b",
"hf": "microsoft/phi-4"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 9053114464,
"minRamBytes": 10411081634,
"minVramBytes": 10411081634,
"sha256": "fd7b6731c33c57f61767612f56517460ec2d1e2e5a3f0163e0eb3d8d8cb5df20"
}
],
"benchmarkProxy": 0.64
},
{
"id": "llama3.3:70b",
"family": "llama3.3",
"params": "70B",
"architecture": "dense",
"license": "llama-3.3-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools",
"reasoning"
],
"releaseDate": "2024-12-06",
"source": {
"ollama": "llama3.3:70b",
"hf": "meta-llama/Llama-3.3-70B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 42520398528,
"minRamBytes": 48898458308,
"minVramBytes": 48898458308,
"sha256": "4824460d29f2058aaf6e1118a63a7a197a09bed509f0e7d4e2efb1ee273b447d"
}
],
"kvBytesPerToken": 327680,
"benchmarkProxy": 0.78
},
{
"id": "olmo2:13b",
"family": "olmo2",
"params": "13B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-11-26",
"source": {
"ollama": "olmo2:13b",
"hf": "allenai/OLMo-2-1124-13B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 8354349536,
"minRamBytes": 9607501967,
"minVramBytes": 9607501967,
"sha256": "cd836509a1a051178be134eba84115eb3a6653a1bd58473a706bf8ee4ab3a764"
}
],
"benchmarkProxy": 0.54
},
{
"id": "olmo2:7b",
"family": "olmo2",
"params": "7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat"
],
"releaseDate": "2024-11-26",
"source": {
"ollama": "olmo2:7b",
"hf": "allenai/OLMo-2-1124-7B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4472020256,
"minRamBytes": 5142823295,
"minVramBytes": 5142823295,
"sha256": "ea89e3927d5ef671159a1359a22cdd418856c4baa2098e665f1c6eed59973968"
}
],
"benchmarkProxy": 0.46
},
{
"id": "qwen2.5-coder:32b",
"family": "qwen2.5-coder",
"params": "32B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"code",
"chat",
"tools"
],
"releaseDate": "2024-11-12",
"source": {
"ollama": "qwen2.5-coder:32b",
"hf": "Qwen/Qwen2.5-Coder-32B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 19851336384,
"minRamBytes": 22829036842,
"minVramBytes": 22829036842,
"sha256": "ac3d1ba8aa77755dab3806d9024e9c385ea0d5b412d6bdf9157f8a4a7e9fc0d9"
}
],
"kvBytesPerToken": 262144,
"benchmarkProxy": 0.72
},
{
"id": "qwen2.5-coder:7b",
"family": "qwen2.5-coder",
"params": "7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"code",
"chat",
"tools"
],
"releaseDate": "2024-11-12",
"source": {
"ollama": "qwen2.5-coder:7b",
"hf": "Qwen/Qwen2.5-Coder-7B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4683074048,
"minRamBytes": 5385535156,
"minVramBytes": 5385535156,
"sha256": "60e05f2100071479f596b964f89f510f057ce397ea22f2833a0cfe029bfc2463"
}
],
"kvBytesPerToken": 57344,
"benchmarkProxy": 0.52
},
{
"id": "smollm2:1.7b",
"family": "smollm2",
"params": "1.7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-11-01",
"source": {
"ollama": "smollm2:1.7b",
"hf": "HuggingFaceTB/SmolLM2-1.7B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1820414656,
"minRamBytes": 2093476855,
"minVramBytes": 2093476855,
"sha256": "4d2396b16114669389d7555c15a1592aad584750310f648edad5ca8c4eccda17"
}
],
"benchmarkProxy": 0.2
},
{
"id": "smollm2:135m",
"family": "smollm2",
"params": "135M",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat"
],
"releaseDate": "2024-11-01",
"source": {
"ollama": "smollm2:135m",
"hf": "HuggingFaceTB/SmolLM2-135M-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 270885952,
"minRamBytes": 311518845,
"minVramBytes": 311518845,
"sha256": "f535f83ec568d040f88ddc04a199fa6da90923bbb41d4dcaed02caa924d6ef57"
}
],
"benchmarkProxy": 0.05
},
{
"id": "smollm2:360m",
"family": "smollm2",
"params": "360M",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat"
],
"releaseDate": "2024-11-01",
"source": {
"ollama": "smollm2:360m",
"hf": "HuggingFaceTB/SmolLM2-360M-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 725553792,
"minRamBytes": 834386861,
"minVramBytes": 834386861,
"sha256": "7d23be4097d67c5c43c3df62ebc19609a25f483c7d0b77d3989e1d94e36ccab6"
}
],
"benchmarkProxy": 0.05
},
{
"id": "granite3-moe:3b",
"family": "granite3-moe",
"params": "3B",
"architecture": "moe",
"activeParams": "0.8B",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-10-21",
"source": {
"ollama": "granite3-moe:3b",
"hf": "ibm-granite/granite-3.0-3b-a800m-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2059355712,
"minRamBytes": 2368259069,
"minVramBytes": 2368259069,
"sha256": "2c931bb3b330f26b3d4bd0ee56b828196750c244b3c40450dfc8e4074962e215"
}
],
"benchmarkProxy": 0.3
},
{
"id": "llama3.2:1b",
"family": "llama3.2",
"params": "1B",
"architecture": "dense",
"license": "llama-3.2-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-09-25",
"source": {
"ollama": "llama3.2:1b",
"hf": "meta-llama/Llama-3.2-1B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1321082688,
"minRamBytes": 1519245092,
"minVramBytes": 1519245092,
"sha256": "74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45"
}
],
"kvBytesPerToken": 32768,
"benchmarkProxy": 0.2
},
{
"id": "llama3.2:3b",
"family": "llama3.2",
"params": "3B",
"architecture": "dense",
"license": "llama-3.2-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-09-25",
"source": {
"ollama": "llama3.2:3b",
"hf": "meta-llama/Llama-3.2-3B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2019377376,
"minRamBytes": 2322283983,
"minVramBytes": 2322283983,
"sha256": "dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff"
}
],
"kvBytesPerToken": 114688,
"benchmarkProxy": 0.28
},
{
"id": "qwen2.5:0.5b",
"family": "qwen2.5",
"params": "0.5B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:0.5b",
"hf": "Qwen/Qwen2.5-0.5B-Instruct",
"gguf": {
"repo": "Qwen/Qwen2.5-0.5B-Instruct-GGUF",
"revision": "9217f5db79a29953eb74d5343926648285ec7e67",
"file": "qwen2.5-0.5b-instruct-q4_k_m.gguf",
"sha256": "74a4da8c9fdbcd15bd1f6d01d621410d31c6fc00986f5eb687824e7b93d7a9db"
}
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 397807936,
"minRamBytes": 457479127,
"minVramBytes": 457479127,
"sha256": "c5396e06af294bd101b30dce59131a76d2b773e76950acc870eda801d3ab0515"
}
],
"kvBytesPerToken": 12288,
"benchmarkProxy": 0.18
},
{
"id": "qwen2.5:0.5b-mlx",
"family": "qwen2.5",
"params": "0.5B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat"
],
"releaseDate": "2024-09-19",
"source": {
"hf": "mlx-community/Qwen2.5-0.5B-Instruct-4bit",
"mlx": {
"repo": "mlx-community/Qwen2.5-0.5B-Instruct-4bit",
"revision": "a5339a4131f135d0fdc6a5c8b5bbed2753bbe0f3",
"files": [
{
"file": "config.json",
"sha256": "b045e57ea90b8f1b35f89f954b176a5c1faa02bd0af2c89bcec191239d66cef4",
"bytes": 783
},
{
"file": "model.safetensors",
"sha256": "ddffab9cbc7bf6dde941c6724841eeca8981fcfa81ca20ff8efff1396326d153",
"bytes": 278064920
},
{
"file": "model.safetensors.index.json",
"sha256": "54001cb4c11197119c206dde28e7be08e5872aab6c6d271aed339ec77e84f870",
"bytes": 44209
},
{
"file": "tokenizer.json",
"sha256": "a8506e7111b80c6d8635951a02eab0f4e1a8e4e5772da83846579e97b16f61bf",
"bytes": 7031673
},
{
"file": "tokenizer_config.json",
"sha256": "f7c61e32b7a17d19bf8e7037dcb74079a833e53ea9801f24008cac68458f03b7",
"bytes": 7308
},
{
"file": "special_tokens_map.json",
"sha256": "76862e765266b85aa9459767e33cbaf13970f327a0e88d1c65846c2ddd3a1ecd",
"bytes": 613
},
{
"file": "vocab.json",
"sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910",
"bytes": 2776833
},
{
"file": "merges.txt",
"sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5",
"bytes": 1671853
},
{
"file": "added_tokens.json",
"sha256": "58b54bbe36fc752f79a24a271ef66a0a0830054b4dfad94bde757d851968060b",
"bytes": 605
}
]
}
},
"quantizations": [
{
"name": "4bit",
"diskBytes": 289598797,
"minRamBytes": 333038617,
"minVramBytes": 333038617
}
],
"kvBytesPerToken": 12288,
"benchmarkProxy": 0.18
},
{
"id": "qwen2.5:1.5b",
"family": "qwen2.5",
"params": "1.5B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:1.5b",
"hf": "Qwen/Qwen2.5-1.5B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 986048512,
"minRamBytes": 1133955789,
"minVramBytes": 1133955789,
"sha256": "183715c435899236895da3869489cc30ac241476b4971a20285b1a462818a5b4"
}
],
"kvBytesPerToken": 28672,
"benchmarkProxy": 0.26
},
{
"id": "qwen2.5:14b",
"family": "qwen2.5",
"params": "14B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:14b",
"hf": "Qwen/Qwen2.5-14B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 8988110688,
"minRamBytes": 10336327292,
"minVramBytes": 10336327292,
"sha256": "2049f5674b1e92b4464e5729975c9689fcfbf0b0e4443ccf10b5339f370f9a54"
}
],
"kvBytesPerToken": 196608,
"benchmarkProxy": 0.58
},
{
"id": "qwen2.5:32b",
"family": "qwen2.5",
"params": "32B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:32b",
"hf": "Qwen/Qwen2.5-32B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 19851336288,
"minRamBytes": 22829036732,
"minVramBytes": 22829036732,
"sha256": "eabc98a9bcbfce7fd70f3e07de599f8fda98120fefed5881934161ede8bd1a41"
}
],
"kvBytesPerToken": 262144,
"benchmarkProxy": 0.7
},
{
"id": "qwen2.5:3b",
"family": "qwen2.5",
"params": "3B",
"architecture": "dense",
"license": "qwen",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:3b",
"hf": "Qwen/Qwen2.5-3B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1929903008,
"minRamBytes": 2219388460,
"minVramBytes": 2219388460,
"sha256": "5ee4f07cdb9beadbbb293e85803c569b01bd37ed059d2715faa7bb405f31caa6"
}
],
"kvBytesPerToken": 36864,
"benchmarkProxy": 0.34
},
{
"id": "qwen2.5:72b",
"family": "qwen2.5",
"params": "72B",
"architecture": "dense",
"license": "qwen",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:72b",
"hf": "Qwen/Qwen2.5-72B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 47415715200,
"minRamBytes": 54528072480,
"minVramBytes": 54528072480,
"sha256": "6e7fdda508e91cb0f63de5c15ff79ac63a1584ccafd751c07ca12b7f442101b8"
}
],
"kvBytesPerToken": 327680,
"benchmarkProxy": 0.76
},
{
"id": "qwen2.5:7b",
"family": "qwen2.5",
"params": "7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-09-19",
"source": {
"ollama": "qwen2.5:7b",
"hf": "Qwen/Qwen2.5-7B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4683073952,
"minRamBytes": 5385535045,
"minVramBytes": 5385535045,
"sha256": "2bada8a7450677000f678be90653b85d364de7db25eb5ea54136ada5f3933730"
}
],
"kvBytesPerToken": 57344,
"benchmarkProxy": 0.5
},
{
"id": "phi3.5:mini",
"family": "phi3.5",
"params": "3.8B",
"architecture": "dense",
"license": "mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-08-20",
"source": {
"ollama": "phi3.5:3.8b",
"hf": "microsoft/Phi-3.5-mini-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2176177152,
"minRamBytes": 2567375000,
"minVramBytes": 2567375000,
"sha256": "b5374915da534cb93df39f03bd4f2cd5a0c533df0d5e21957dc9556c260be9eb"
}
],
"benchmarkProxy": 0.36
},
{
"id": "gemma2:2b",
"family": "gemma2",
"params": "2B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat"
],
"releaseDate": "2024-07-31",
"source": {
"ollama": "gemma2:2b",
"hf": "google/gemma-2-2b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 1629509152,
"minRamBytes": 1873935525,
"minVramBytes": 1873935525,
"sha256": "7462734796d67c40ecec2ca98eddf970e171dbb6b370e43fd633ee75b69abe1b"
}
],
"benchmarkProxy": 0.32
},
{
"id": "llama3.1:70b",
"family": "llama3.1",
"params": "70B",
"architecture": "dense",
"license": "llama-3.1-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-07-23",
"source": {
"ollama": "llama3.1:70b",
"hf": "meta-llama/Llama-3.1-70B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 42520398176,
"minRamBytes": 48898457903,
"minVramBytes": 48898457903,
"sha256": "de20d2cf2dc430b1717a8b07a9df029d651f3895dbffec4729a3902a6fe344c9"
}
],
"kvBytesPerToken": 327680,
"benchmarkProxy": 0.74
},
{
"id": "llama3.1:8b",
"family": "llama3.1",
"params": "8B",
"architecture": "dense",
"license": "llama-3.1-community",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-07-23",
"source": {
"ollama": "llama3.1:8b",
"hf": "meta-llama/Llama-3.1-8B-Instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4920738944,
"minRamBytes": 5658849786,
"minVramBytes": 5658849786,
"sha256": "667b0c1932bc6ffc593ed1d03f895bf2dc8dc6df21db3042284a6f4416b06a29"
}
],
"kvBytesPerToken": 131072,
"benchmarkProxy": 0.48
},
{
"id": "mistral-nemo:12b",
"family": "mistral-nemo",
"params": "12B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-07-18",
"source": {
"ollama": "mistral-nemo:12b",
"hf": "mistralai/Mistral-Nemo-Instruct-2407"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 7071700672,
"minRamBytes": 8132455773,
"minVramBytes": 8132455773,
"sha256": "b559938ab7a0392fc9ea9675b82280f2a15669ec3e0e0fc491c9cb0a7681cf94"
}
],
"kvBytesPerToken": 163840,
"benchmarkProxy": 0.55
},
{
"id": "gemma2:27b",
"family": "gemma2",
"params": "27B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-06-27",
"source": {
"ollama": "gemma2:27b",
"hf": "google/gemma-2-27b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 15628378336,
"minRamBytes": 18241875000,
"minVramBytes": 18241875000,
"sha256": "d7e4b00a7d7a8d03d4eed9b0f3f61a427e9f0fc5dea6aeb414e41dee23dc8ecc"
}
],
"benchmarkProxy": 0.68
},
{
"id": "gemma2:9b",
"family": "gemma2",
"params": "9B",
"architecture": "dense",
"license": "gemma",
"openWeight": true,
"contextLength": 8192,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-06-27",
"source": {
"ollama": "gemma2:9b",
"hf": "google/gemma-2-9b-it"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5443143296,
"minRamBytes": 6259614791,
"minVramBytes": 6259614791,
"sha256": "ff1d1fc78170d787ee1201778e2dd65ea211654ca5fb7d69b5a2e7b123a50373"
}
],
"benchmarkProxy": 0.48
},
{
"id": "yi:34b",
"family": "yi",
"params": "34B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-05-12",
"source": {
"ollama": "yi:34b",
"hf": "01-ai/Yi-1.5-34B-Chat"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 19466529344,
"minRamBytes": 22971250000,
"minVramBytes": 22971250000,
"sha256": "648f108f6a1ef7e12c33003d2308028ec98356d4ebda894d7db541f9e8c11aeb"
}
],
"benchmarkProxy": 0.67
},
{
"id": "yi:6b",
"family": "yi",
"params": "6B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat"
],
"releaseDate": "2024-05-12",
"source": {
"ollama": "yi:6b",
"hf": "01-ai/Yi-1.5-6B-Chat"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 3479326560,
"minRamBytes": 4053750000,
"minVramBytes": 4053750000,
"sha256": "3502ee168c97b13ce3966acea684973e73c22f0ede62ca05f886b009b43b2f09"
}
],
"benchmarkProxy": 0.39
},
{
"id": "yi:9b",
"family": "yi",
"params": "9B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 4096,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-05-12",
"source": {
"ollama": "yi:9b",
"hf": "01-ai/Yi-1.5-9B-Chat"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 5036994816,
"minRamBytes": 6080625000,
"minVramBytes": 6080625000,
"sha256": "59b1e2a78cb0a5b8f7ba25e459ee17788242cd996e1f62a295dcbd65e5344b11"
}
],
"benchmarkProxy": 0.47
},
{
"id": "phi3:mini",
"family": "phi3",
"params": "3.8B",
"architecture": "dense",
"license": "mit",
"openWeight": true,
"contextLength": 131072,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2024-04-23",
"source": {
"ollama": "phi3:mini",
"hf": "microsoft/Phi-3-mini-128k-instruct"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 2176177120,
"minRamBytes": 2567375000,
"minVramBytes": 2567375000,
"sha256": "633fc5be925f9a484b61d6f9b9a78021eeb462100bd557309f01ba84cac26adf"
}
],
"benchmarkProxy": 0.34
},
{
"id": "mixtral:8x22b",
"family": "mixtral",
"params": "141B",
"architecture": "moe",
"activeParams": "39B",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 65536,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2024-04-17",
"source": {
"ollama": "mixtral:8x22b",
"hf": "mistralai/Mixtral-8x22B-Instruct-v0.1"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 79521266688,
"minRamBytes": 95263125000,
"minVramBytes": 95263125000,
"sha256": "94e70a590f93b8f12eb2bc6ddb2e48b3f2fab0bc577fee4b1cb42305ba590916"
}
],
"benchmarkProxy": 0.83
},
{
"id": "mixtral:8x7b",
"family": "mixtral",
"params": "47B",
"architecture": "moe",
"activeParams": "13B",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"code",
"tools"
],
"releaseDate": "2023-12-11",
"source": {
"ollama": "mixtral:8x7b",
"hf": "mistralai/Mixtral-8x7B-Instruct-v0.1"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 26443590560,
"minRamBytes": 31754375000,
"minVramBytes": 31754375000,
"sha256": "f2dc41fa964b42bfe34e9fb09c0acdcfbfd6e52f1332930b4eacc9d6ad1c6cd2"
}
],
"benchmarkProxy": 0.75
},
{
"id": "mistral:7b",
"family": "mistral",
"params": "7B",
"architecture": "dense",
"license": "apache-2.0",
"openWeight": true,
"contextLength": 32768,
"capabilities": [
"chat",
"code"
],
"releaseDate": "2023-09-27",
"source": {
"ollama": "mistral:7b",
"hf": "mistralai/Mistral-7B-Instruct-v0.3"
},
"quantizations": [
{
"name": "Q4_K_M",
"diskBytes": 4372811712,
"minRamBytes": 5028733469,
"minVramBytes": 5028733469,
"sha256": "f5074b1221da0f5a2910d33b642efa5b9eb58cfdddca1c79e16d7ad28aa2b31f"
}
],
"kvBytesPerToken": 131072,
"benchmarkProxy": 0.46
}
]
}