Hardware / NVIDIA /H100 80GB SXM

NVIDIA · GPU

H100 80GB SXM

Runs 154 models fully in memory, 9 more with CPU offload. 4 of those sit close enough to 80 GB that our size estimate is what decides them.

Memory

80 GB

Bandwidth

3350 GB/s

Runs

154

01

Specifications

Memory

80 GB

Decides which models fit at all.

Memory bandwidth

3350 GB/s

How fast it can read the model. Sets the speed of the reply.

Compute

989 TFLOPS

Trillions of half-precision sums a second. Sets how fast it reads a long prompt, not how fast it replies.

Power draw

700 W

At full load, for sizing a power supply.

MSRP

$30,000

What the maker asked for it new. We hold no used or street prices, so this is not what one costs today.

Released

Oct 1, 2022

Architecture

Hopper

02

What it runs

Best quant per model. Text models first, then speech, then what does not fit.

What is quantisation? →

ModelParamsContextSpeedVerdict

Granite 4.1 8B8.8B131K287 tok/sFits in memory Qwen3.6 35B A3B36B262K1253 tok/sFits in memory Gemma 4 26B A4B25.8B131K801 tok/sFits in memory Qwen3.5-9B9.7B262K260 tok/sFits in memory Qwen3.5-27B27.8B131K159 tok/sFits in memory GLM 4.7 Flash31.2B131K120 tok/sFits in memory Nemotron 3 Nano 30B A3B31.6B262K1253 tok/sFits in memory Ministral 3 14B 251213.9B262K182 tok/sFits in memory Ministral 3 8B 25128.9B262K423 tok/sFits in memory Ministral 3 3B 25123.8B131K664 tok/sFits in memory Olmo 3 32B Think32.2B66K137 tok/sFits in memory gpt-oss-safeguard-20b21.5B131K890 tok/sFits in memory Qwen3 VL 32B Instruct33.4B131K132 tok/sFits in memory Granite 4.0 Micro3.2B66K1178 tok/sFits in memory Qwen3 VL 8B Thinking8.8B131K502 tok/sFits in memory Qwen3 VL 30B A3B Thinking31.1B262K1253 tok/sFits in memory Cydonia 24B V4.123.6B131K160 tok/sFits in memory Qwen3 Next 80B A3B Thinking81.3B131K1068 tok/sFits in memory UI-TARS 7B8.3B66K454 tok/sFits in memory Uncensored24B66K184 tok/sFits in memory Hunyuan A13B Instruct80.4B66K246 tok/sFits in memory Mistral Small 3.2 24B24B131K184 tok/sFits in memory Llama Guard 4 12B12B262K314 tok/sFits in memory Qwen3 14B14.8B131K170 tok/sFits in memory Gemma 3 12B12.2B131K362 tok/sFits in memory Gemma 3 27B27.4B262K161 tok/sFits in memory Rocinante 12B12.2B66K207 tok/sFits in memory Llama 3.2 3B Instruct3.2B131K788 tok/sFits in memory Mistral Nemo12.2B131K362 tok/sFits in memory Qwen3.6 27B27.8B131K136 tok/sFits in memory Qwen3.5-35B-A3B36B262K1253 tok/sFits in memory Qwen3 VL 8B Instruct8.8B262K428 tok/sFits in memory Qwen3 VL 30B A3B Instruct31.1B262K1253 tok/sFits in memory Qwen3 Next 80B A3B Instruct81.3B131K1253 tok/sFits in memory Qwen3 30B A3B Thinking 250730.5B66K715 tok/sFits in memory Qwen3 Coder 30B A3B Instruct30.5B262K715 tok/sFits in memory Qwen3 30B A3B Instruct 250730.5B262K1253 tok/sFits in memory Qwen3 8B8.2B131K539 tok/sFits in memory Qwen3 32B32.8B131K135 tok/sFits in memory Mistral Small 3.1 24B24B66K157 tok/sFits in memory Gemma 3 4B4.3B131K876 tok/sFits in memory Reka Flash 320.9B66K212 tok/sFits in memory Llama 3.1 8B Instruct8B131K553 tok/sFits in memory Muse Glimmer 30B29.8B131K126 tok/sFits in memory Nemotron 3.5 Lightning31.6B262K119 tok/sFits in memory Qwen3.8 27B27.8B131K136 tok/sFits in memory Granite 4.2 8B8.8B131K287 tok/sFits in memory Nemotron 3.5 Content Safety4.3B131K876 tok/sFits in memory Ternary Bonsai 2 27B27B262K140 tok/sFits in memory Laguna XS 2.133.4B262K113 tok/sFits in memory Nex-N2-Mini35.1B262K107 tok/sFits in memory Gemma 4 31B31.3B33K141 tok/sFits in memory Gemma 3n 4B7.8B33K483 tok/sFits in memory Skyfall 36B V236.9B33K120 tok/sFits in memory Qwen2.5 Coder 32B Instruct32.8B33K135 tok/sFits in memory UnslopNemo 12B12.2B33K309 tok/sFits in memory Llama 3.2 1B Instruct1.2B33K3684 tok/sFits in memory Mistral Small 323.6B33K187 tok/sFits in memory Qwen2.5 7B Instruct7.6B33K332 tok/sFits in memory Nex-N2.5-Mini35.1B262K107 tok/sFits in memory Reka Edge7.1B16K355 tok/sFits in memory Voxtral Small 24B 250724.3B16K155 tok/sFits in memory GLM 4.5 Air111B16K313 tok/sFits in memoryest Phi 414.7B16K256 tok/sFits in memory Llama 3 8B Lunaris8B8K471 tok/sFits in memory MythoMax 13B13B8K290 tok/sFits in memory Gemma 2 27B27.2B8K163 tok/sFits in memory Hy-MT2-1.8B2B8K1879 tok/sFits in memory Hy-MT2-30B-A3B30.1B8K1068 tok/sFits in memory Hy-MT2-7B8B8K400 tok/sFits in memory ReMM SLERP 13B13B4K340 tok/sFits in memory Hermes 4 70B70.6B66K63 tok/sFits in memory Llama 3.3 Euryale 70B70.6B66K63 tok/sFits in memory Llama 3.1 Euryale 70B v2.270.6B66K63 tok/sFits in memory Hermes 3 70B Instruct70.6B66K63 tok/sFits in memory Llama 3.1 70B Instruct70.6B131K63 tok/sFits in memory Llama 3.3 70B Instruct70.6B131K63 tok/sFits in memory Qwen2.5 VL 72B Instruct73.4B66K60 tok/sFits in memory Qwen3 Coder Next79.7B131K47 tok/sFits in memory Qwen2.5 72B Instruct72.7B33K61 tok/sFits in memory Magnum v4 72B72.7B16K61 tok/sFits in memory Llama 4 Scout109B33K41 tok/sFits in memoryest R1 Distill Llama 70B70.6B8K63 tok/sFits in memory GLM 4.6V108B33K35 tok/sFits in memory GLM 4.5V108B33K35 tok/sFits in memory Command A111B16K40 tok/sFits in memoryest Ling-2.6-flash107B8K35 tok/sFits in memoryest

Speech models it runs67 modelsTranscription and speech models that fit in memory. Throughput figures for a transcription model come from its text decoder, so treat this as a fit answer rather than a speed one.

Niagara 19m Batch.en20M221031 tok/sFits in memory Niagara 38m Batch.en38M66378 tok/sFits in memory Higgs Audio v3 8b STT v28.9B497 tok/sFits in memory Higgs Audio v3 STT2.7B934 tok/sFits in memory Cohere Transcribe 03 20262.1B1201 tok/sFits in memory Distil Large v3.50.8B3153 tok/sFits in memory Lite Whisper Large v3 Acc1.4B2691 tok/sFits in memory Owsm CTC v3.1 1B1.1B4019 tok/sFits in memory Owsm CTC v3.2 ft 1B1B2522 tok/sFits in memory Owsm CTC v4 1B1B4421 tok/sFits in memory Hubert Large Ls960 ft0.3B7882 tok/sFits in memory MMS 1b All1B3768 tok/sFits in memory Wav2vec2 Large 960h Lv60 Self0.3B7882 tok/sFits in memory Hojo ASR V15.2B727 tok/sFits in memory Granite 4.0 1b Speech2.3B1097 tok/sFits in memory Granite Speech 3.3 2b3B1474 tok/sFits in memory Granite Speech 3.3 8b8.6B514 tok/sFits in memory Granite Speech 4.1 2b2.3B1922 tok/sFits in memory Granite Speech 4.1 2b NAR2.3B1922 tok/sFits in memory STT 2.6b en2.6B1700 tok/sFits in memory Phi 4 Multimodal Instruct5.6B131K789 tok/sFits in memory VibeVoice ASR HF8.3B304 tok/sFits in memory Voxtral Mini 3B 25074.7B537 tok/sFits in memory Voxtral Mini 4B Realtime 26024.4B573 tok/sFits in memory Canary 180m Flash0.2B20933 tok/sFits in memory Canary 1b1B2522 tok/sFits in memory Canary 1b Flash0.8B3153 tok/sFits in memory Canary 1b v21B2522 tok/sFits in memory Canary Qwen 2.5b2.6B970 tok/sFits in memory Nemotron 3.5 ASR Streaming 0.6b0.6B4204 tok/sFits in memory Nemotron Speech Streaming en 0.6b0.6B4204 tok/sFits in memory Parakeet CTC 0.6b0.6B7368 tok/sFits in memory Parakeet CTC 1.1b1.1B2293 tok/sFits in memory Parakeet RNNT 0.6b0.6B4204 tok/sFits in memory Parakeet RNNT 1.1b1.1B4019 tok/sFits in memory Parakeet TDT CTC 110m0.1B22931 tok/sFits in memory Parakeet TDT 0.6b v20.6B6280 tok/sFits in memory Parakeet TDT 0.6b v30.6B4204 tok/sFits in memory Parakeet TDT 1.1b1.1B2293 tok/sFits in memory Nyra Health CrisperWhisper1.6B1577 tok/sFits in memory Whisper Large v31.5B2512 tok/sFits in memory Whisper Large v3 Turbo0.8B5526 tok/sFits in memory MOSS Transcribe Diarize0.9B2803 tok/sFits in memory Qwen3 ASR 0.6B0.9B4912 tok/sFits in memory Qwen3 ASR 0.6B HF0.8B5526 tok/sFits in memory Qwen3 ASR 1.7B2.3B1638 tok/sFits in memory Qwen3 ASR 1.7B HF2B1261 tok/sFits in memory Zipformer cr CTC Transducer XL 290M0.3B12993 tok/sFits in memory Zipformer Transducer XL 290M0.3B8698 tok/sFits in memory ASR Conformer Largescaleasr0.5B9210 tok/sFits in memory GLM ASR Nano 25122.3B1922 tok/sFits in memory Granite Speech 5.0 470m Turboctc0.5B7536 tok/sFits in memory Granite Speech 5.0 470m Turboctc nc0.5B8841 tok/sFits in memory Fun ASR Nano 2512 HF0.8B5526 tok/sFits in memory Gemma 4 12B it12B314 tok/sFits in memory Gemma 4 E2B it5.1B739 tok/sFits in memory Gemma 4 E4B it8B131K315 tok/sFits in memory Niagara 9m Batch.en9M280261 tok/sFits in memory Niagara 84m Batch.en84M52626 tok/sFits in memory ARK ASR 0.6B1.3B33K2899 tok/sFits in memory ARK ASR 3B4.1B33K1078 tok/sFits in memory Audio8 ASR 0.1B0.3B33K8408 tok/sFits in memory MOSS Transcribe Preview 2B2.4B33K1842 tok/sFits in memory Moonshine Streaming Medium0.3B4K14735 tok/sFits in memory Moonshine Streaming Small0.1B4K25224 tok/sFits in memory Moonshine Streaming Tiny0B4Kno speed estimateFits in memory Moonshine Tiny0B194no speed estimateFits in memory

Runs only by spilling into system memory9 modelsThese load, but part of the weights sits in ordinary system memory, which is far slower than the chip.

Laguna S 2.1118B—no speed estimateSpills to system RAM Mistral Medium 3.5128B—no speed estimateSpills to system RAM Mistral Small 4119B—no speed estimateSpills to system RAM Nemotron 3 Super124B—no speed estimateSpills to system RAM Qwen3.5-122B-A10B125B—no speed estimateSpills to system RAM Devstral 2 2512125B—no speed estimateSpills to system RAM gpt-oss-120b120B—no speed estimateSpills to system RAM Mixtral 8x22B Instruct141B—no speed estimateSpills to system RAMest Ling-3.0-flash128B—no speed estimateSpills to system RAMest

Too large for this device61 modelsThe weights do not fit even with part of them offloaded to system memory. Closest calls first.

Step 3.7 Flash201B—no speed estimateToo large Step 3.5 Flash199B—no speed estimateToo large MiniMax M2.5229B—no speed estimateToo large Qwen3 235B A22B Thinking 2507235B—no speed estimateToo large Qwen3 VL 235B A22B Thinking236B—no speed estimateToo large Qwen3 VL 235B A22B Instruct236B—no speed estimateToo large DeepSeek V4 Flash291B—no speed estimateToo large Hy3299B—no speed estimateToo large GLM 5.3 Flash321B—no speed estimateToo large GLM 4.5358B—no speed estimateToo large Laguna M.1226B—no speed estimateToo large Nex-N2-Pro397B—no speed estimateToo large Nex-N2.5-Pro397B—no speed estimateToo large Trinity Large Thinking399B—no speed estimateToo large MiniMax M2.7229B—no speed estimateToo large MiniMax M2.1229B—no speed estimateToo large MiniMax M2229B—no speed estimateToo large Llama 4 Maverick402B—no speed estimateToo large Qwen3 235B A22B Instruct 2507235B—no speed estimateToo large GLM 4.6357B—no speed estimateToo large GLM 4.7358B—no speed estimateToo large ERNIE 4.5 VL 424B A47B424B—no speed estimateToo large Inkling Small266B—no speed estimateToo large MiniMax M3427B—no speed estimateToo large Hy3 preview299B—no speed estimateToo large DeepSeek V4 Flash Vision Exp305B—no speed estimateToo large MiniMax M1456B—no speed estimateToo large MiniMax-01456B—no speed estimateToo large MiMo-V2.5311B—no speed estimateToo large Nemotron 3 Ultra561B—no speed estimateToo large Qwen3 Coder 480B A35B480B—no speed estimateToo large R1 0528685B—no speed estimateToo large DeepSeek V3 0324685B—no speed estimateToo large DeepSeek V3.1685B—no speed estimateToo large Jamba Large 1.7399B—no speed estimateToo large Qwen3.5 397B A17B403B—no speed estimateToo large Hermes 4 405B406B—no speed estimateToo large Hermes 3 405B Instruct406B—no speed estimateToo large GLM 5.1754B—no speed estimateToo large GLM 5.2753B—no speed estimateToo large DeepSeek V3685B—no speed estimateToo large GLM 5754B—no speed estimateToo large DeepSeek V4.1 Flash763B—no speed estimateToo large MiMo-V2.5-Pro1T—no speed estimateToo large Kimi K2 09051T—no speed estimateToo large Kimi K2 07111T—no speed estimateToo large Ling-2.6-1T1T—no speed estimateToo large DeepSeek V3.1 Terminus685B—no speed estimateToo large DeepSeek V3.2685B—no speed estimateToo large DeepSeek V3.2 Exp685B—no speed estimateToo large Ring-2.6-1T1T—no speed estimateToo large Kimi K2.7 Code1.1T—no speed estimateToo large Kimi K2.61.1T—no speed estimateToo large Kimi K2.51.1T—no speed estimateToo large Hy4 preview780B—no speed estimateToo large Inkling952B—no speed estimateToo large Kimi K2 Thinking1.1T—no speed estimateToo large LongCat 2.01.8T—no speed estimateToo large DeepSeek V4 Pro1.6T—no speed estimateToo large Qwen3.8 2.4T A95B2.4T—no speed estimateToo large Kimi K32.8T—no speed estimateToo large

224 models have a verdict on this device: 154 fit in memory, 9 spill into system memory and 61 are too large. Every one of them is on this page.

03

Other Hopper cards

The same generation in other memory sizes. Memory is what decides which models fit.

H200 141GB SXM141 GB · more memory

Something wrong on this page? Tell us