Which model for 8 × B300 · 2304 Go?
2304 GB usable for the model. 121 open-weight models fit in memory.
The margin goes to context: count ~1 GB per 8k tokens for a 7B model. A small margin means a short context.
Unlocked by this setup
These models fit on no smaller setup in the catalogue (the next one down offers 1440 usable GB), even in their lightest quantization.
| Model | Elo | Quantization | RAM required | Margin | Memory headroom | Run |
|---|---|---|---|---|---|---|
| Kimi K3Moonshot AI · 2800B | 1487 | Q4_K_M | 1758 Go | +546 Go | tight | Hugging Face ↗ |
| Kimi K3.1 ProMoonshot AI · 2800B | 1485 | MXFP4 | 1718 Go | +586 Go | tight | Hugging Face ↗ |
Also fit on a smaller setup
| Model | Elo | Quantization | RAM required | Margin | Memory headroom | Run |
|---|---|---|---|---|---|---|
| GLM-5.3Zhipu AI · 753B | 1482 | FP16 | 1659 Go | +645 Go | tight | Hugging Face ↗ |
| DeepSeek V4.1-FlashDeepSeek · 552B | 1475 | FP8 | 563 Go | +1741 Go | comfortable | Hugging Face ↗ |
| GLM-5.2Zhipu AI · 754B | 1475 | FP16 | 1661 Go | +643 Go | tight | Hugging Face ↗ |
| GLM-5Zhipu AI · 744B | 1457 | FP16 | 1639 Go | +665 Go | tight | Hugging Face ↗ |
| Gemma 4 31BGoogle · 31B | 1452 | FP16 | 70 Go | +2234 Go | comfortable | Hugging Face ↗ |
| Qwen3.5 397B-A17BAlibaba · 397B | 1441 | FP16 | 875 Go | +1429 Go | comfortable | Hugging Face ↗ |
| DeepSeek V4-FlashDeepSeek · 284B | 1439 | Q4_K_MFP16 also fits · 627 Go | 180 Go | +2124 Go | comfortable | ollama run deepseek-v4-flash |
| Gemma 4 26B-A4BGoogle · 26B | 1437 | FP16 | 59 Go | +2245 Go | comfortable | Hugging Face ↗ |
| Qwen3.8 27BAlibaba · 27B | 1434 | FP16 | 61 Go | +2243 Go | comfortable | Hugging Face ↗ |
| Mistral Medium 3.5Mistral AI · 128B | 1426 | FP16 | 284 Go | +2020 Go | comfortable | Hugging Face ↗ |
| Nemotron 3 Ultra 550B-A55BNVIDIA · 550B | 1426 | FP16 | 1212 Go | +1092 Go | comfortable | Hugging Face ↗ |
| Mistral Large 4Mistral AI · 1000B | 1420 | FP8 | 1168 Go | +1136 Go | comfortable | Hugging Face ↗ |
| Kimi K2Moonshot AI · 1026B | 1418 | FP16 | 2259 Go | +45 Go | tight | Hugging Face ↗ |
| Mistral Large 3Mistral AI · 675B | 1415 | FP16 | 1487 Go | +817 Go | comfortable | Hugging Face ↗ |
| GLM-4.5Z.ai · 355B | 1411 | FP16 | 783 Go | +1521 Go | comfortable | Hugging Face ↗ |
| DeepSeek-R1DeepSeek · 671B | 1398 | Q4_K_MFP16 also fits · 1478 Go | 423 Go | +1881 Go | comfortable | ollama run deepseek-r1:671b |
| DeepSeek V3DeepSeek · 685B | 1395 | Q4_K_MFP16 also fits · 1509 Go | 431 Go | +1873 Go | comfortable | ollama run deepseek-v3 |
| Qwen3-Coder 480BAlibaba · 480B | 1387 | Q4_K_MFP16 also fits · 1058 Go | 303 Go | +2001 Go | comfortable | ollama run qwen3-coder:480b |
| Qwen3 235B-A22BAlibaba · 235B | 1375 | Q4_K_MFP16 also fits · 519 Go | 149 Go | +2155 Go | comfortable | ollama run qwen3:235b |
| GLM-4.5 AirZ.ai · 106B | 1373 | FP16 | 235 Go | +2069 Go | comfortable | Hugging Face ↗ |
| Gemma 3 27BGoogle · 27B | 1365 | Q4_K_MFP16 also fits · 61 Go | 19 Go | +2285 Go | comfortable | ollama run gemma3:27b |
| Nemotron 3 Super 120B-A12BNVIDIA · 120B | 1360 | FP16 | 266 Go | +2038 Go | comfortable | Hugging Face ↗ |
| Mistral Small 3.2Mistral AI · 24B | 1356 | Q4_K_MFP16 also fits · 55 Go | 17 Go | +2287 Go | comfortable | ollama run mistral-small3.2 |
| gpt-oss-120bOpenAI · 117B | 1352 | MXFP4 | 73 Go | +2231 Go | comfortable | ollama run gpt-oss:120b |
| Qwen3 32BAlibaba · 32.8B | 1347 | Q4_K_MFP16 also fits · 74 Go | 23 Go | +2281 Go | comfortable | ollama run qwen3:32b |
| Gemma 3 12BGoogle · 12B | 1342 | Q4_K_MFP16 also fits · 28 Go | 9 Go | +2295 Go | comfortable | ollama run gemma3:12b |
| QwQ 32BAlibaba · 32.8B | 1336 | Q4_K_MFP16 also fits · 74 Go | 23 Go | +2281 Go | comfortable | ollama run qwq |
| Llama 4 MaverickMeta · 400B | 1327 | Q4_K_MFP16 also fits · 882 Go | 253 Go | +2051 Go | comfortable | ollama run llama4:maverick |
| Llama 4 ScoutMeta · 109B | 1322 | Q4_K_MFP16 also fits · 242 Go | 70 Go | +2234 Go | comfortable | ollama run llama4:scout |
| Gemma 3n E4BGoogle · 4B | 1318 | Q4_K_MFP16 also fits · 11 Go | 5 Go | +2299 Go | comfortable | ollama run gemma3n:e4b |
| gpt-oss-20bOpenAI · 21B | 1318 | MXFP4 | 17 Go | +2287 Go | comfortable | ollama run gpt-oss:20b |
| Llama 3.3 70BMeta · 70.6B | 1318 | Q4_K_MFP16 also fits · 157 Go | 46 Go | +2258 Go | comfortable | ollama run llama3.3:70b |
| Nemotron 3 Nano 30B-A3BNVIDIA · 30B | 1314 | FP16 | 68 Go | +2236 Go | comfortable | Hugging Face ↗ |
| Granite 4.1 8BIBM · 8B | 1304 | FP16 | 20 Go | +2284 Go | comfortable | Hugging Face ↗ |
| Gemma 3 4BGoogle · 4B | 1303 | Q4_K_MFP16 also fits · 11 Go | 5 Go | +2299 Go | comfortable | ollama run gemma3:4b |
| Qwen2.5 72BAlibaba · 72.7B | 1303 | Q4_K_MFP16 also fits · 162 Go | 48 Go | +2256 Go | comfortable | ollama run qwen2.5:72b |
| Llama 3.1 Nemotron 70BNVIDIA · 70.6B | 1299 | Q4_K_MFP16 also fits · 157 Go | 46 Go | +2258 Go | comfortable | ollama run nemotron:70b |
| Qwen2.5-Coder 32BAlibaba · 32.8B | 1271 | Q4_K_MFP16 also fits · 74 Go | 23 Go | +2281 Go | comfortable | ollama run qwen2.5-coder:32b |
| Aya Expanse 32BCohere · 32.3B | 1267 | Q4_K_MFP16 also fits · 73 Go | 22 Go | +2282 Go | comfortable | ollama run aya-expanse:32b |
| Command R+Cohere · 104B | 1262 | Q4_K_MFP16 also fits · 231 Go | 67 Go | +2237 Go | comfortable | ollama run command-r-plus |
| Phi-4Microsoft · 14.7B | 1256 | Q4_K_MFP16 also fits · 34 Go | 11 Go | +2293 Go | comfortable | ollama run phi4 |
| Phi-4 ReasoningMicrosoft · 14.7B | 1256 | Q4_K_MFP16 also fits · 34 Go | 11 Go | +2293 Go | comfortable | ollama run phi4-reasoning |
| OLMo 2 32BAllen AI · 32B | 1252 | FP16 | 72 Go | +2232 Go | comfortable | Hugging Face ↗ |
| CodestralMistral AI · 22.2B | 1240 | Q4_K_MFP16 also fits · 51 Go | 16 Go | +2288 Go | comfortable | ollama run codestral |
| Ministral 8BMistral AI · 8.02B | 1238 | FP16 | 20 Go | +2284 Go | comfortable | Hugging Face ↗ |
| Llama 3.1 8BMeta · 8.03B | 1211 | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run llama3.1:8b |
| Mixtral 8x7BMistral AI · 46.7B | 1197 | Q4_K_MFP16 also fits · 105 Go | 31 Go | +2273 Go | comfortable | ollama run mixtral:8x7b |
| Llama 3.2 3BMeta · 3.21B | 1167 | Q4_K_MFP16 also fits · 9 Go | 4 Go | +2300 Go | comfortable | ollama run llama3.2:3b |
| Llama 3.2 1BMeta · 1.24B | 1111 | Q4_K_MFP16 also fits · 5 Go | 3 Go | +2301 Go | comfortable | ollama run llama3.2:1b |
| Mistral 7BMistral AI · 7.25B | 1111 | Q4_K_MFP16 also fits · 18 Go | 7 Go | +2297 Go | comfortable | ollama run mistral:7b |
Not yet ranked by the Arena
No Arena Elo linked: recent release, model absent from the Arena, or a match we have not made yet.
| Model | Elo | Quantization | RAM required | Margin | Memory headroom | Run |
|---|---|---|---|---|---|---|
| Qwen3 0.6BAlibaba · 0.6B | — | Q4_K_MFP16 also fits · 3 Go | 2 Go | +2302 Go | comfortable | ollama run qwen3:0.6b |
| Gemma 3 1BGoogle · 1B | — | Q4_K_MFP16 also fits · 4 Go | 3 Go | +2301 Go | comfortable | ollama run gemma3:1b |
| Phi-4 MiniMicrosoft · 3.8B | — | Q4_K_MFP16 also fits · 10 Go | 4 Go | +2300 Go | comfortable | ollama run phi4-mini |
| Qwen3 4BAlibaba · 4.02B | — | Q4_K_MFP16 also fits · 11 Go | 5 Go | +2299 Go | comfortable | ollama run qwen3:4b |
| Gemma 4 E2BGoogle · 2B | — | FP16 | 6 Go | +2298 Go | comfortable | Hugging Face ↗ |
| CodeGemma 7BGoogle · 8.5B | — | Q4_K_MFP16 also fits · 21 Go | 7 Go | +2297 Go | comfortable | ollama run codegemma |
| Command R7BCohere · 8B | — | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run command-r7b |
| DeepSeek-R1 Distill 8BDeepSeek · 8.03B | — | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run deepseek-r1:8b |
| Granite 3.3 8BIBM · 8.1B | — | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run granite3.3:8b |
| Hermes 3 8BNous Research · 8.03B | — | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run hermes3:8b |
| MiniCPM5-2BOpenBMB · 2.5B | — | FP16 | 7 Go | +2297 Go | comfortable | Hugging Face ↗ |
| Qwen3 8BAlibaba · 8.2B | — | Q4_K_MFP16 also fits · 20 Go | 7 Go | +2297 Go | comfortable | ollama run qwen3:8b |
| Falcon 3 10BTII · 10.3B | — | Q4_K_MFP16 also fits · 25 Go | 8 Go | +2296 Go | comfortable | ollama run falcon3:10b |
| Granite 4.1 3BIBM · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| Ministral 3 3BMistral AI · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| SmolLM3 3BHugging Face · 3.1B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| Mistral Nemo 12BMistral AI · 12.2B | — | Q4_K_MFP16 also fits · 29 Go | 10 Go | +2294 Go | comfortable | ollama run mistral-nemo |
| Gemma 4 E4BGoogle · 4B | — | FP16 | 11 Go | +2293 Go | comfortable | Hugging Face ↗ |
| Qwen3 14BAlibaba · 14.8B | — | Q4_K_MFP16 also fits · 35 Go | 11 Go | +2293 Go | comfortable | ollama run qwen3:14b |
| DeepSeek-Coder-V2 LiteDeepSeek · 15.7B | — | Q4_K_MFP16 also fits · 37 Go | 12 Go | +2292 Go | comfortable | ollama run deepseek-coder-v2:16b |
| StarCoder 2BigCode · 16B | — | Q4_K_MFP16 also fits · 37 Go | 12 Go | +2292 Go | comfortable | ollama run starcoder2:15b |
| Devstral SmallMistral AI · 24B | — | Q4_K_MFP16 also fits · 55 Go | 17 Go | +2287 Go | comfortable | ollama run devstral |
| Lucie 7BOpenLLM France · 6.7B | — | FP16 | 17 Go | +2287 Go | comfortable | Hugging Face ↗ |
| Magistral SmallMistral AI · 24B | — | Q4_K_MFP16 also fits · 55 Go | 17 Go | +2287 Go | comfortable | ollama run magistral |
| OLMo 3 7BAllen AI · 7B | — | FP16 | 17 Go | +2287 Go | comfortable | Hugging Face ↗ |
| Apertus 8BSwiss AI · 8B | — | FP16 | 20 Go | +2284 Go | comfortable | Hugging Face ↗ |
| Ministral 3 8BMistral AI · 8B | — | FP16 | 20 Go | +2284 Go | comfortable | Hugging Face ↗ |
| InternLM3 8BShanghai AI Lab · 8.8B | — | FP16 | 21 Go | +2283 Go | comfortable | Hugging Face ↗ |
| Apertus 1.5 8BSwiss AI Initiative · 9B | — | FP16 | 22 Go | +2282 Go | comfortable | Hugging Face ↗ |
| EuroLLM 9BUTTER Project · 9.15B | — | FP16 | 22 Go | +2282 Go | comfortable | Hugging Face ↗ |
| DeepSeek-R1 Distill 32BDeepSeek · 32.8B | — | Q4_K_MFP16 also fits · 74 Go | 23 Go | +2281 Go | comfortable | ollama run deepseek-r1:32b |
| Gemma 4 12BGoogle · 12B | — | FP16 | 28 Go | +2276 Go | comfortable | Hugging Face ↗ |
| Ministral 3 14BMistral AI · 14B | — | FP16 | 33 Go | +2271 Go | comfortable | Hugging Face ↗ |
| Qwen3.6 27BAlibaba · 27B | — | FP16 | 61 Go | +2243 Go | comfortable | Hugging Face ↗ |
| Qwen3.6 35B-A3BAlibaba · 35B | — | FP16 | 79 Go | +2225 Go | comfortable | Hugging Face ↗ |
| Apertus 1.5 70BSwiss AI Initiative · 70B | — | FP16 | 156 Go | +2148 Go | comfortable | Hugging Face ↗ |
| Kolibri 1Aleph Alpha · 78.1B | — | FP16 | 174 Go | +2130 Go | comfortable | Hugging Face ↗ |
| Mistral Small 4Mistral AI · 119B | — | FP16 | 264 Go | +2040 Go | comfortable | Hugging Face ↗ |
| Kimi K2.7 CodeMoonshot AI · 1000B | — | FP16 | 2202 Go | +102 Go | tight | Hugging Face ↗ |
Image, video, audio and 3D
These models do not generate text, so the text Arena does not rank them. Indicative order, not a quality ranking.
| Model | Elo | Quantization | RAM required | Margin | Memory headroom | Run |
|---|---|---|---|---|---|---|
| Kokoro-82MHexgrad · 0.08B | — | FP16 | 2 Go | +2302 Go | comfortable | Hugging Face ↗ |
| ChatterboxResemble AI · 0.5B | — | FP16 | 3 Go | +2301 Go | comfortable | Hugging Face ↗ |
| Parakeet TDT 0.6B v3NVIDIA · 0.6B | — | FP16 | 3 Go | +2301 Go | comfortable | Hugging Face ↗ |
| TripoSRStability AI · 0.5B | — | FP16 | 3 Go | +2301 Go | comfortable | Hugging Face ↗ |
| Stable Fast 3DStability AI · 1B | — | FP16 | 4 Go | +2300 Go | comfortable | Hugging Face ↗ |
| TRELLISMicrosoft · 1.2B | — | FP16 | 5 Go | +2299 Go | comfortable | Hugging Face ↗ |
| Whisper large-v3OpenAI · 1.55B | — | FP16 | 5 Go | +2299 Go | comfortable | Hugging Face ↗ |
| Dia 1.6BNari Labs · 1.6B | — | FP16 | 6 Go | +2298 Go | comfortable | Hugging Face ↗ |
| Hunyuan3D 2.1Tencent · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| Orpheus 3BCanopy Labs · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| Voxtral Mini 3BMistral AI · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| YuE2-3Bm-a-p (Multimodal Art Projection) · 3B | — | FP16 | 9 Go | +2295 Go | comfortable | Hugging Face ↗ |
| FLUX.2 [klein] 4BBlack Forest Labs · 4B | — | FP16 | 11 Go | +2293 Go | comfortable | Hugging Face ↗ |
| TRELLIS.2Microsoft · 4B | — | FP16 | 11 Go | +2293 Go | comfortable | Hugging Face ↗ |
| CogVideoX 5BZhipu AI · 5B | — | FP16 | 13 Go | +2291 Go | comfortable | Hugging Face ↗ |
| Wan 2.2 5BAlibaba · 5B | — | FP16 | 13 Go | +2291 Go | comfortable | Hugging Face ↗ |
| Ming-Image 0.1 DesignAnt Group · 6B | — | FP16 | 15 Go | +2289 Go | comfortable | Hugging Face ↗ |
| NitroFusion2SketchX (University of Surrey) · 6B | — | FP16 | 15 Go | +2289 Go | comfortable | Hugging Face ↗ |
| Z-Image TurboAlibaba · 6B | — | FP16 | 15 Go | +2289 Go | comfortable | Hugging Face ↗ |
| HunyuanVideo 1.5Tencent · 8.3B | — | FP16 | 20 Go | +2284 Go | comfortable | Hugging Face ↗ |
| GLM-ImageZhipu AI · 9B | — | FP16 | 22 Go | +2282 Go | comfortable | Hugging Face ↗ |
| Mochi 1Genmo · 10B | — | FP16 | 24 Go | +2280 Go | comfortable | Hugging Face ↗ |
| FLUX.1 [dev]Black Forest Labs · 12B | — | FP16 | 28 Go | +2276 Go | comfortable | Hugging Face ↗ |
| HiDream-I1HiDream · 17B | — | FP16 | 39 Go | +2265 Go | comfortable | Hugging Face ↗ |
| Qwen-Image 2512Alibaba · 20B | — | FP16 | 46 Go | +2258 Go | comfortable | Hugging Face ↗ |
| LTX-2.3Lightricks · 22B | — | FP16 | 50 Go | +2254 Go | comfortable | Hugging Face ↗ |
| Voxtral Small 24BMistral AI · 24B | — | FP16 | 55 Go | +2249 Go | comfortable | Hugging Face ↗ |
| Wan 2.2 A14BAlibaba · 27B | — | FP16 | 61 Go | +2243 Go | comfortable | Hugging Face ↗ |
| FLUX.2 [dev]Black Forest Labs · 32B | — | FP16 | 72 Go | +2232 Go | comfortable | Hugging Face ↗ |
| HunyuanImage 3.0Tencent · 80B | — | FP16 | 178 Go | +2126 Go | comfortable | Hugging Face ↗ |
An “ollama run” command without a suffix pulls the Q4_K_M build from the Ollama library: the row then shows that quantization and its memory, and flags whether FP16 or Q8_0 would also fit. Exception: a model published in a native format (MXFP4 for gpt-oss, FP8 for DeepSeek V4.1) exists in one build only, which is what the command loads. via Hugging Face or an explicit Ollama tag.
Share this verdict
One link, the same shortlist, the same reasons.