LIVE
OpenAI moves toward a rigorous lattice definition of the Standard Model15/11/26 · OpenAI|Open-Weight Reasoning Model Learns to Operate Real Robots Through Simulation-Only Training13/11/26|Forensic analysis finds traces of export-controlled GPUs in Chinese frontier models12/11/26|Google DeepMind unveils Gemini 4 Hydrogen12/11/26 · Google DeepMind|AlphaGenome 2: DeepMind refines its model for decoding non-coding DNA11/11/26 · Google DeepMind|Google DeepMind rolls out Gemini 4 Carbon10/11/26 · Google DeepMind|Google DeepMind unveils AlphaProof Code for formal software verification10/11/26 · Google DeepMind|OpenAI launches a dedicated business presence feature in ChatGPT10/11/26 · OpenAI|Anthropic launches investor roadshow ahead of planned IPO10/11/26 · Anthropic|A petition calls for Jensen Huang's return as Nvidia CEO16/11/26 · Nvidia|OpenAI strengthens tools against online child sexual exploitation16/11/26 · OpenAI|OpenAI Expands Free ChatGPT Access to 200,000 Researchers16/11/26 · OpenAI|OpenAI moves toward a rigorous lattice definition of the Standard Model15/11/26 · OpenAI|Open-Weight Reasoning Model Learns to Operate Real Robots Through Simulation-Only Training13/11/26|Forensic analysis finds traces of export-controlled GPUs in Chinese frontier models12/11/26|Google DeepMind unveils Gemini 4 Hydrogen12/11/26 · Google DeepMind|AlphaGenome 2: DeepMind refines its model for decoding non-coding DNA11/11/26 · Google DeepMind|Google DeepMind rolls out Gemini 4 Carbon10/11/26 · Google DeepMind|Google DeepMind unveils AlphaProof Code for formal software verification10/11/26 · Google DeepMind|OpenAI launches a dedicated business presence feature in ChatGPT10/11/26 · OpenAI|Anthropic launches investor roadshow ahead of planned IPO10/11/26 · Anthropic|A petition calls for Jensen Huang's return as Nvidia CEO16/11/26 · Nvidia|OpenAI strengthens tools against online child sexual exploitation16/11/26 · OpenAI|OpenAI Expands Free ChatGPT Access to 200,000 Researchers16/11/26 · OpenAI|
Hardware Match

Which model for 8 × B200 · 1440 Go?

1440 GB usable for the model. 119 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 1128 usable GB), even in their lightest quantization.

ModelEloQuantizationRAM requiredMarginMemory headroomRun
Mistral Large 4Mistral AI · 1000B1420FP81168 Go+272 GotightHugging Face ↗

Also fit on a smaller setup

ModelEloQuantizationRAM requiredMarginMemory headroomRun
GLM-5.3Zhipu AI · 753B1482Q8_0888 Go+552 GocomfortableHugging Face ↗
DeepSeek V4.1-FlashDeepSeek · 552B1475FP8563 Go+877 GocomfortableHugging Face ↗
GLM-5.2Zhipu AI · 754B1475Q8_0889 Go+551 GocomfortableHugging Face ↗
GLM-5Zhipu AI · 744B1457Q8_0878 Go+562 GocomfortableHugging Face ↗
Gemma 4 31BGoogle · 31B1452FP1670 Go+1370 GocomfortableHugging Face ↗
Qwen3.5 397B-A17BAlibaba · 397B1441FP16875 Go+565 GocomfortableHugging Face ↗
DeepSeek V4-FlashDeepSeek · 284B1439Q4_K_MFP16 also fits · 627 Go180 Go+1260 Gocomfortable ollama run deepseek-v4-flash
Gemma 4 26B-A4BGoogle · 26B1437FP1659 Go+1381 GocomfortableHugging Face ↗
Qwen3.8 27BAlibaba · 27B1434FP1661 Go+1379 GocomfortableHugging Face ↗
Mistral Medium 3.5Mistral AI · 128B1426FP16284 Go+1156 GocomfortableHugging Face ↗
Nemotron 3 Ultra 550B-A55BNVIDIA · 550B1426FP161212 Go+228 GotightHugging Face ↗
Kimi K2Moonshot AI · 1026B1418Q8_01210 Go+230 GotightHugging Face ↗
Mistral Large 3Mistral AI · 675B1415Q8_0796 Go+644 GocomfortableHugging Face ↗
GLM-4.5Z.ai · 355B1411FP16783 Go+657 GocomfortableHugging Face ↗
DeepSeek-R1DeepSeek · 671B1398Q4_K_MQ8_0 also fits · 792 Go423 Go+1017 Gocomfortable ollama run deepseek-r1:671b
DeepSeek V3DeepSeek · 685B1395Q4_K_MQ8_0 also fits · 808 Go431 Go+1009 Gocomfortable ollama run deepseek-v3
Qwen3-Coder 480BAlibaba · 480B1387Q4_K_MFP16 also fits · 1058 Go303 Go+1137 Gocomfortable ollama run qwen3-coder:480b
Qwen3 235B-A22BAlibaba · 235B1375Q4_K_MFP16 also fits · 519 Go149 Go+1291 Gocomfortable ollama run qwen3:235b
GLM-4.5 AirZ.ai · 106B1373FP16235 Go+1205 GocomfortableHugging Face ↗
Gemma 3 27BGoogle · 27B1365Q4_K_MFP16 also fits · 61 Go19 Go+1421 Gocomfortable ollama run gemma3:27b
Nemotron 3 Super 120B-A12BNVIDIA · 120B1360FP16266 Go+1174 GocomfortableHugging Face ↗
Mistral Small 3.2Mistral AI · 24B1356Q4_K_MFP16 also fits · 55 Go17 Go+1423 Gocomfortable ollama run mistral-small3.2
gpt-oss-120bOpenAI · 117B1352MXFP473 Go+1367 Gocomfortable ollama run gpt-oss:120b
Qwen3 32BAlibaba · 32.8B1347Q4_K_MFP16 also fits · 74 Go23 Go+1417 Gocomfortable ollama run qwen3:32b
Gemma 3 12BGoogle · 12B1342Q4_K_MFP16 also fits · 28 Go9 Go+1431 Gocomfortable ollama run gemma3:12b
QwQ 32BAlibaba · 32.8B1336Q4_K_MFP16 also fits · 74 Go23 Go+1417 Gocomfortable ollama run qwq
Llama 4 MaverickMeta · 400B1327Q4_K_MFP16 also fits · 882 Go253 Go+1187 Gocomfortable ollama run llama4:maverick
Llama 4 ScoutMeta · 109B1322Q4_K_MFP16 also fits · 242 Go70 Go+1370 Gocomfortable ollama run llama4:scout
Gemma 3n E4BGoogle · 4B1318Q4_K_MFP16 also fits · 11 Go5 Go+1435 Gocomfortable ollama run gemma3n:e4b
gpt-oss-20bOpenAI · 21B1318MXFP417 Go+1423 Gocomfortable ollama run gpt-oss:20b
Llama 3.3 70BMeta · 70.6B1318Q4_K_MFP16 also fits · 157 Go46 Go+1394 Gocomfortable ollama run llama3.3:70b
Nemotron 3 Nano 30B-A3BNVIDIA · 30B1314FP1668 Go+1372 GocomfortableHugging Face ↗
Granite 4.1 8BIBM · 8B1304FP1620 Go+1420 GocomfortableHugging Face ↗
Gemma 3 4BGoogle · 4B1303Q4_K_MFP16 also fits · 11 Go5 Go+1435 Gocomfortable ollama run gemma3:4b
Qwen2.5 72BAlibaba · 72.7B1303Q4_K_MFP16 also fits · 162 Go48 Go+1392 Gocomfortable ollama run qwen2.5:72b
Llama 3.1 Nemotron 70BNVIDIA · 70.6B1299Q4_K_MFP16 also fits · 157 Go46 Go+1394 Gocomfortable ollama run nemotron:70b
Qwen2.5-Coder 32BAlibaba · 32.8B1271Q4_K_MFP16 also fits · 74 Go23 Go+1417 Gocomfortable ollama run qwen2.5-coder:32b
Aya Expanse 32BCohere · 32.3B1267Q4_K_MFP16 also fits · 73 Go22 Go+1418 Gocomfortable ollama run aya-expanse:32b
Command R+Cohere · 104B1262Q4_K_MFP16 also fits · 231 Go67 Go+1373 Gocomfortable ollama run command-r-plus
Phi-4Microsoft · 14.7B1256Q4_K_MFP16 also fits · 34 Go11 Go+1429 Gocomfortable ollama run phi4
Phi-4 ReasoningMicrosoft · 14.7B1256Q4_K_MFP16 also fits · 34 Go11 Go+1429 Gocomfortable ollama run phi4-reasoning
OLMo 2 32BAllen AI · 32B1252FP1672 Go+1368 GocomfortableHugging Face ↗
CodestralMistral AI · 22.2B1240Q4_K_MFP16 also fits · 51 Go16 Go+1424 Gocomfortable ollama run codestral
Ministral 8BMistral AI · 8.02B1238FP1620 Go+1420 GocomfortableHugging Face ↗
Llama 3.1 8BMeta · 8.03B1211Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run llama3.1:8b
Mixtral 8x7BMistral AI · 46.7B1197Q4_K_MFP16 also fits · 105 Go31 Go+1409 Gocomfortable ollama run mixtral:8x7b
Llama 3.2 3BMeta · 3.21B1167Q4_K_MFP16 also fits · 9 Go4 Go+1436 Gocomfortable ollama run llama3.2:3b
Llama 3.2 1BMeta · 1.24B1111Q4_K_MFP16 also fits · 5 Go3 Go+1437 Gocomfortable ollama run llama3.2:1b
Mistral 7BMistral AI · 7.25B1111Q4_K_MFP16 also fits · 18 Go7 Go+1433 Gocomfortable 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.

ModelEloQuantizationRAM requiredMarginMemory headroomRun
Qwen3 0.6BAlibaba · 0.6B—Q4_K_MFP16 also fits · 3 Go2 Go+1438 Gocomfortable ollama run qwen3:0.6b
Gemma 3 1BGoogle · 1B—Q4_K_MFP16 also fits · 4 Go3 Go+1437 Gocomfortable ollama run gemma3:1b
Phi-4 MiniMicrosoft · 3.8B—Q4_K_MFP16 also fits · 10 Go4 Go+1436 Gocomfortable ollama run phi4-mini
Qwen3 4BAlibaba · 4.02B—Q4_K_MFP16 also fits · 11 Go5 Go+1435 Gocomfortable ollama run qwen3:4b
Gemma 4 E2BGoogle · 2B—FP166 Go+1434 GocomfortableHugging Face ↗
CodeGemma 7BGoogle · 8.5B—Q4_K_MFP16 also fits · 21 Go7 Go+1433 Gocomfortable ollama run codegemma
Command R7BCohere · 8B—Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run command-r7b
DeepSeek-R1 Distill 8BDeepSeek · 8.03B—Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run deepseek-r1:8b
Granite 3.3 8BIBM · 8.1B—Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run granite3.3:8b
Hermes 3 8BNous Research · 8.03B—Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run hermes3:8b
MiniCPM5-2BOpenBMB · 2.5B—FP167 Go+1433 GocomfortableHugging Face ↗
Qwen3 8BAlibaba · 8.2B—Q4_K_MFP16 also fits · 20 Go7 Go+1433 Gocomfortable ollama run qwen3:8b
Falcon 3 10BTII · 10.3B—Q4_K_MFP16 also fits · 25 Go8 Go+1432 Gocomfortable ollama run falcon3:10b
Granite 4.1 3BIBM · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
Ministral 3 3BMistral AI · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
SmolLM3 3BHugging Face · 3.1B—FP169 Go+1431 GocomfortableHugging Face ↗
Mistral Nemo 12BMistral AI · 12.2B—Q4_K_MFP16 also fits · 29 Go10 Go+1430 Gocomfortable ollama run mistral-nemo
Gemma 4 E4BGoogle · 4B—FP1611 Go+1429 GocomfortableHugging Face ↗
Qwen3 14BAlibaba · 14.8B—Q4_K_MFP16 also fits · 35 Go11 Go+1429 Gocomfortable ollama run qwen3:14b
DeepSeek-Coder-V2 LiteDeepSeek · 15.7B—Q4_K_MFP16 also fits · 37 Go12 Go+1428 Gocomfortable ollama run deepseek-coder-v2:16b
StarCoder 2BigCode · 16B—Q4_K_MFP16 also fits · 37 Go12 Go+1428 Gocomfortable ollama run starcoder2:15b
Devstral SmallMistral AI · 24B—Q4_K_MFP16 also fits · 55 Go17 Go+1423 Gocomfortable ollama run devstral
Lucie 7BOpenLLM France · 6.7B—FP1617 Go+1423 GocomfortableHugging Face ↗
Magistral SmallMistral AI · 24B—Q4_K_MFP16 also fits · 55 Go17 Go+1423 Gocomfortable ollama run magistral
OLMo 3 7BAllen AI · 7B—FP1617 Go+1423 GocomfortableHugging Face ↗
Apertus 8BSwiss AI · 8B—FP1620 Go+1420 GocomfortableHugging Face ↗
Ministral 3 8BMistral AI · 8B—FP1620 Go+1420 GocomfortableHugging Face ↗
InternLM3 8BShanghai AI Lab · 8.8B—FP1621 Go+1419 GocomfortableHugging Face ↗
Apertus 1.5 8BSwiss AI Initiative · 9B—FP1622 Go+1418 GocomfortableHugging Face ↗
EuroLLM 9BUTTER Project · 9.15B—FP1622 Go+1418 GocomfortableHugging Face ↗
DeepSeek-R1 Distill 32BDeepSeek · 32.8B—Q4_K_MFP16 also fits · 74 Go23 Go+1417 Gocomfortable ollama run deepseek-r1:32b
Gemma 4 12BGoogle · 12B—FP1628 Go+1412 GocomfortableHugging Face ↗
Ministral 3 14BMistral AI · 14B—FP1633 Go+1407 GocomfortableHugging Face ↗
Qwen3.6 27BAlibaba · 27B—FP1661 Go+1379 GocomfortableHugging Face ↗
Qwen3.6 35B-A3BAlibaba · 35B—FP1679 Go+1361 GocomfortableHugging Face ↗
Apertus 1.5 70BSwiss AI Initiative · 70B—FP16156 Go+1284 GocomfortableHugging Face ↗
Kolibri 1Aleph Alpha · 78.1B—FP16174 Go+1266 GocomfortableHugging Face ↗
Mistral Small 4Mistral AI · 119B—FP16264 Go+1176 GocomfortableHugging Face ↗
Kimi K2.7 CodeMoonshot AI · 1000B—Q8_01179 Go+261 GotightHugging 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.

ModelEloQuantizationRAM requiredMarginMemory headroomRun
Kokoro-82MHexgrad · 0.08B—FP162 Go+1438 GocomfortableHugging Face ↗
ChatterboxResemble AI · 0.5B—FP163 Go+1437 GocomfortableHugging Face ↗
Parakeet TDT 0.6B v3NVIDIA · 0.6B—FP163 Go+1437 GocomfortableHugging Face ↗
TripoSRStability AI · 0.5B—FP163 Go+1437 GocomfortableHugging Face ↗
Stable Fast 3DStability AI · 1B—FP164 Go+1436 GocomfortableHugging Face ↗
TRELLISMicrosoft · 1.2B—FP165 Go+1435 GocomfortableHugging Face ↗
Whisper large-v3OpenAI · 1.55B—FP165 Go+1435 GocomfortableHugging Face ↗
Dia 1.6BNari Labs · 1.6B—FP166 Go+1434 GocomfortableHugging Face ↗
Hunyuan3D 2.1Tencent · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
Orpheus 3BCanopy Labs · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
Voxtral Mini 3BMistral AI · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
YuE2-3Bm-a-p (Multimodal Art Projection) · 3B—FP169 Go+1431 GocomfortableHugging Face ↗
FLUX.2 [klein] 4BBlack Forest Labs · 4B—FP1611 Go+1429 GocomfortableHugging Face ↗
TRELLIS.2Microsoft · 4B—FP1611 Go+1429 GocomfortableHugging Face ↗
CogVideoX 5BZhipu AI · 5B—FP1613 Go+1427 GocomfortableHugging Face ↗
Wan 2.2 5BAlibaba · 5B—FP1613 Go+1427 GocomfortableHugging Face ↗
Ming-Image 0.1 DesignAnt Group · 6B—FP1615 Go+1425 GocomfortableHugging Face ↗
NitroFusion2SketchX (University of Surrey) · 6B—FP1615 Go+1425 GocomfortableHugging Face ↗
Z-Image TurboAlibaba · 6B—FP1615 Go+1425 GocomfortableHugging Face ↗
HunyuanVideo 1.5Tencent · 8.3B—FP1620 Go+1420 GocomfortableHugging Face ↗
GLM-ImageZhipu AI · 9B—FP1622 Go+1418 GocomfortableHugging Face ↗
Mochi 1Genmo · 10B—FP1624 Go+1416 GocomfortableHugging Face ↗
FLUX.1 [dev]Black Forest Labs · 12B—FP1628 Go+1412 GocomfortableHugging Face ↗
HiDream-I1HiDream · 17B—FP1639 Go+1401 GocomfortableHugging Face ↗
Qwen-Image 2512Alibaba · 20B—FP1646 Go+1394 GocomfortableHugging Face ↗
LTX-2.3Lightricks · 22B—FP1650 Go+1390 GocomfortableHugging Face ↗
Voxtral Small 24BMistral AI · 24B—FP1655 Go+1385 GocomfortableHugging Face ↗
Wan 2.2 A14BAlibaba · 27B—FP1661 Go+1379 GocomfortableHugging Face ↗
FLUX.2 [dev]Black Forest Labs · 32B—FP1672 Go+1368 GocomfortableHugging Face ↗
HunyuanImage 3.0Tencent · 80B—FP16178 Go+1262 GocomfortableHugging 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.