Data center GPU · NVIDIA
What AI models can an H200 SXM 141GB run?
141 GB is server territory. The constraint stops being memory and becomes bandwidth: everything fits, and what varies is how long you wait for it. At 4800 GB/s there is enough bandwidth to keep generation responsive for anything that fits.
The short answer
Assuming an 8k context window and default settings, these are the models worth downloading first.
Best all-rounder
MiniMax M2.7
Q4_K_M · 130.55 GB · about 253 tokens/s
Best for code
Llama 3.3 70B Instruct
Q8_0 · 71.81 GB · about 49.6 tokens/s
Largest that still runs well
Qwen3 235B A22B
235.09B parameters · Q4_K_M · about 147.5 tokens/s
Fastest useful answer
LFM2.5 8B A1B
about 1243.4 tokens/s · 8.81 GB
See how fast it feels
H200 SXM 141GB running MiniMax M2.7 at Q4_K_M
YouWhy does my model use more memory when the conversation gets longer?
Reading your question
Simulated from our estimate at a 512-token question, not a recording. Assumes nothing else is competing for the GPU.
Where it sits
With 141 GB it belongs to the 48 GB tier. The 48 GB guide explains what that tier runs and where it runs out, and the bandwidth guide explains why 4800 GB/s decides how fast it answers.
We do not track a price for this device, so it does not appear in the budget tools.
Every model, scored on this device
Each row uses the highest-quality quantization that both fits and stays conversational. Speed is a single-stream estimate at 8k context.
| Model | Params | Verdict | Download | Memory used | Speed | Max context |
|---|---|---|---|---|---|---|
| MiniMax M2.7 MiniMax | 228.69B (10.98B active) | Runs great | Q4_K_M | 130.55 GB | 253 t/s | 32k |
| Qwen3.5 122B A10B Qwen | 125.09B (8.17B active) | Runs great | Q8_0 | 123.73 GB | 244.8 t/s | 256k |
| GPT OSS 120B GPT-OSS | 116.83B (5.7B active) | Runs great | MXFP4 | 59.4 GB | 649.9 t/s | 128k |
| MiniMax M2.5 MiniMax | 228.7B (10.98B active) | Runs great | Q4_K_M | 130.56 GB | 253 t/s | 32k |
| Llama 3.3 70B Instruct Llama | 70.55B | Runs great | Q8_0 | 71.81 GB | 49.6 t/s | 128k |
| Qwen3 235B A22B Qwen | 235.09B (22.14B active) | Runs great | Q4_K_M | 133.71 GB | 147.5 t/s | 32k |
| Qwen2.5 VL 72B Instruct Qwen | 73.41B | Runs great | Q8_0 | 74.45 GB | 47.8 t/s | 64k |
| Qwen2.5 72B Instruct Qwen | 72.71B | Runs great | Q8_0 | 73.76 GB | 48.2 t/s | 32k |
| Gemma 4 31B Instruct Gemma | 31.27B | Runs great | Q8_0 | 34.67 GB | 103.2 t/s | 256k |
| GLM 4.5 Air GLM | 110.47B (13.4B active) | Runs great | Q8_0 | 110.54 GB | 140.1 t/s | 128k |
| Qwen3 32B Qwen | 32.76B | Runs great | Q8_0 | 34.09 GB | 105 t/s | 32k |
| Qwen3.5 27B Qwen | 27.78B | Runs great | Q8_0 | 27.32 GB | 131.4 t/s | 256k |
| Llama 3 3 Nemotron Super 49B V1 Llama | 49.87B | Runs great | Q8_0 | 68.85 GB | 51.7 t/s | 32k |
| Qwen3 Coder Next Qwen | 79.67B (3.19B active) | Runs great | Q8_0 | 79.13 GB | 601.8 t/s | 256k |
| Granite 4.1 30B Granite | 28.87B | Runs great | Q8_0 | 30.97 GB | 115.5 t/s | 128k |
| Qwen2.5 32B Instruct Qwen | 32.76B | Runs great | Q8_0 | 34.09 GB | 105 t/s | 32k |
| Qwen2.5 Coder 32B Instruct Qwen | 32.76B | Runs great | Q8_0 | 34.09 GB | 105 t/s | 32k |
| Qwen3 Next 80B A3B Instruct Qwen | 81.32B (3.19B active) | Runs great | Q8_0 | 80.76 GB | 601.7 t/s | 256k |
| Gemma 3 27B Instruct Gemma | 27.43B | Runs great | Q8_0 | 28.68 GB | 125.1 t/s | 128k |
| Qwen3.5 35B A3B Qwen | 35.95B (2.9B active) | Runs great | Q8_0 | 35.63 GB | 672.6 t/s | 256k |
| Gemma 4 26B A4B Instruct Gemma | 25.81B (3.74B active) | Runs great | Q8_0 | 26.72 GB | 453.1 t/s | 256k |
| GLM 4.7 Flash GLM | 31.22B (3.66B active) | Runs great | Q8_0 | 31.33 GB | 512.3 t/s | 128k |
| Mixtral 8x7B Instruct V0.1 Mistral | 46.7B (12.88B active) | Runs great | Q8_0 | 47.49 GB | 149.4 t/s | 32k |
| Qwen3 30B A3B Qwen | 30.53B (3.34B active) | Runs great | Q8_0 | 30.99 GB | 509.4 t/s | 32k |
| Mistral Small 24B Instruct 2501 Mistral | 23.57B | Runs great | Q8_0 | 24.35 GB | 147.7 t/s | 32k |
| Phi 3.5 MoE Instruct Phi | 41.87B (6.64B active) | Runs great | Q8_0 | 42.71 GB | 271.3 t/s | 128k |
| Gemma 4 12B Instruct Gemma | 11.96B | Runs great | Q8_0 | 13.55 GB | 268.6 t/s | 256k |
| Qwen3 14B Qwen | 14.77B | Runs great | Q8_0 | 15.54 GB | 233.9 t/s | 32k |
| GPT OSS 20B GPT-OSS | 20.91B (4.18B active) | Runs great | MXFP4 | 11.55 GB | 857.7 t/s | 128k |
| Qwen3 Coder 30B A3B Instruct Qwen | 30.53B (3.34B active) | Runs great | Q8_0 | 30.99 GB | 509.4 t/s | 32k |
Show the other 56 models
| Qwen2.5 14B Instruct | 14.77B | Runs great | Q8_0 | 15.79 GB | 230.1 t/s | 32k |
| Qwen2.5 Coder 14B Instruct | 14.77B | Runs great | Q8_0 | 15.79 GB | 230.1 t/s | 32k |
| Codestral 22B V0.1 | 22.25B | Runs great | Q8_0 | 24.04 GB | 149.8 t/s | 32k |
| Qwen3.5 9B | 9.65B | Runs great | Q8_0 | 9.25 GB | 399.5 t/s | 256k |
| Phi 4 | 14.66B | Runs great | Q8_0 | 16 GB | 227 t/s | 16k |
| Gemma 3 12B Instruct | 12.19B | Runs great | Q8_0 | 13.27 GB | 274.3 t/s | 128k |
| Gemma 4 E4B Instruct | 8B | Runs great | Q8_0 | 8.42 GB | 438.1 t/s | 128k |
| Granite 4.1 8B | 8.79B | Runs great | Q8_0 | 10.35 GB | 355 t/s | 128k |
| Qwen3 8B | 8.19B | Runs great | Q8_0 | 9.02 GB | 410 t/s | 32k |
| Mistral Nemo Instruct 2407 | 12.25B | Runs great | Q8_0 | 13.15 GB | 277.9 t/s | 128k |
| Qwen2.5 VL 7B Instruct | 8.29B | Runs great | Q8_0 | 8.49 GB | 435.9 t/s | 64k |
| Qwen2.5 7B Instruct | 7.62B | Runs great | Q8_0 | 7.83 GB | 474.8 t/s | 32k |
| Llama 3.1 8B Instruct | 8.03B | Runs great | Q8_0 | 8.83 GB | 419.1 t/s | 128k |
| DeepSeek Coder V2 Lite Instruct | 15.71B | Runs great | Q8_0 | 15.91 GB | 703.3 t/s | 128k |
| Gemma 2 9B Instruct | 9.24B | Runs great | Q8_0 | 11.5 GB | 317.8 t/s | 8k |
| Qwen2.5 Coder 7B Instruct | 7.62B | Runs great | Q8_0 | 7.83 GB | 474.8 t/s | 32k |
| OLMo 3 7B Instruct | 7.3B | Runs great | Q8_0 | 9.72 GB | 379 t/s | 64k |
| Internlm3 8B Instruct | 8.8B | Runs great | Q8_0 | 8.97 GB | 412.5 t/s | 32k |
| Mistral 7B Instruct V0.3 | 7.25B | Runs great | Q8_0 | 8.45 GB | 439.2 t/s | 32k |
| Gemma 4 E2B Instruct | 5.12B | Runs great | Q8_0 | 5.44 GB | 689.9 t/s | 128k |
| Qwen3.5 4B | 4.66B | Runs great | Q8_0 | 5.27 GB | 719.4 t/s | 256k |
| Mistral 7B Instruct V0.2 | 7.24B | Runs great | Q8_0 | 8.44 GB | 439.6 t/s | 32k |
| DeepSeek Coder 7B Instruct V1.5 | 6.91B | Runs great | Q8_0 | 10.58 GB | 347.2 t/s | 4k |
| DeepSeek Coder 6.7B Instruct | 6.74B | Runs great | Q8_0 | 10.94 GB | 335.1 t/s | 16k |
| LFM2.5 8B A1B | 8.47B | Runs great | Q8_0 | 8.81 GB | 1243.4 t/s | 64k |
| Qwen3 4B | 4.02B | Runs great | Q8_0 | 5.46 GB | 692.1 t/s | 32k |
| Gemma 3 4B Instruct | 4.3B | Runs great | Q8_0 | 4.91 GB | 775.9 t/s | 128k |
| Phi 4 Mini Instruct | 3.84B | Runs great | Q8_0 | 5.17 GB | 735.8 t/s | 128k |
| Granite 4.1 3B | 3.4B | Runs great | Q8_0 | 4.34 GB | 885.2 t/s | 128k |
| Qwen2.5 3B Instruct | 3.09B | Runs great | Q8_0 | 3.68 GB | 1057.7 t/s | 32k |
| Phi 3 Mini 4k Instruct | 3.82B | Runs great | Q8_0 | 4.8 GB | 796.8 t/s | 4k |
| Llama 3.2 3B Instruct | 3.21B | Runs great | Q8_0 | 4.42 GB | 871.7 t/s | 128k |
| LFM2.5 2.6B | 2.7B | Runs great | Q8_0 | 3.14 GB | 1262.8 t/s | 128k |
| Qwen3.5 2B | 2.27B | Runs great | Q8_0 | 2.7 GB | 1497.6 t/s | 256k |
| Gemma 2 2B Instruct | 2.61B | Runs great | Q8_0 | 3.54 GB | 1106.4 t/s | 8k |
| Starcoder2 3B | 3.03B | Runs great | Q8_0 | 3.34 GB | 1190.7 t/s | 16k |
| Qwen3 1.7B | 2.03B | Runs great | Q8_0 | 3.22 GB | 1224.6 t/s | 32k |
| DeepSeek R1 Distill Qwen 1.5B | 1.78B | Runs great | Q8_0 | 2.07 GB | 2019 t/s | 128k |
| Qwen2.5 1.5B Instruct | 1.54B | Runs great | Q8_0 | 2.06 GB | 2026.6 t/s | 32k |
| SmolLM2 1.7B | 1.71B | Runs great | Q8_0 | 3.53 GB | 1106.3 t/s | 8k |
| OLMo 2 0425 1B | 1.48B | Runs great | Q8_0 | 2.6 GB | 1561.9 t/s | 4k |
| Llama 3.2 1B Instruct | 1.24B | Runs great | Q8_0 | 1.82 GB | 2391 t/s | 128k |
| LFM2.5 1.2B Instruct | 1.17B | Runs great | Q8_0 | 1.46 GB | 3156.9 t/s | 64k |
| Gemma 3 1B Instruct | 1B | Runs great | Q8_0 | 1.35 GB | 3399.5 t/s | 32k |
| Qwen3.5 0.8B | 0.87B | Runs great | Q8_0 | 1.28 GB | 3630.3 t/s | 256k |
| Qwen3 0.6B | 0.75B | Runs great | Q8_0 | 1.92 GB | 2183.8 t/s | 32k |
| Qwen2.5 0.5B Instruct | 0.49B | Runs great | Q8_0 | 0.88 GB | 6103.4 t/s | 32k |
| LFM2.5 350M | 0.35B | Runs great | Q8_0 | 0.68 GB | 9450.8 t/s | 64k |
| GLM 4.5 | 358.34B | Partial offload | n/a | 180.23 GB | 13.3 t/s | n/a |
| MiniMax M3 | 427.04B | Partial offload | n/a | 212.09 GB | 12.4 t/s | n/a |
| DeepSeek V3.2 | 685.4B | Partial offload | n/a | 385.91 GB | 3.8 t/s | n/a |
| DeepSeek R1 | 684.53B | Partial offload | n/a | 385.42 GB | 3.8 t/s | n/a |
| GLM 5.1 | 753.86B | Partial offload | n/a | 424.51 GB | 3.3 t/s | n/a |
| GLM 5.2 | 753.33B | Partial offload | n/a | 424.21 GB | 3.3 t/s | n/a |
| Kimi K2.6 | 1026.88B | Partial offload | n/a | 577.78 GB | 3.2 t/s | n/a |
| Kimi K2 Instruct | 1026.41B | Partial offload | n/a | 577.52 GB | 3.2 t/s | n/a |
What it will not run
0 of the 86 architectures we track are out of reach here, even at 2-bit precision. Another 8 run by splitting the model between the card and system RAM, which works but drops generation to single digits for most dense models.
Questions people ask about an H200 SXM 141GB
How much VRAM does an H200 SXM 141GB have?
The H200 SXM 141GB has 141 GB of VRAM, of which roughly 140.2 GB is available to a model once the operating system takes its share.
Can an H200 SXM 141GB run a 70B model?
Yes. At Q8_0 a 70B model fits in 71.81 GB and generates around 49.6 tokens per second.
What is the best local model for an H200 SXM 141GB?
MiniMax M2.7 at Q4_K_M is the strongest all-round fit: 130.55 GB used, around 253 tokens per second at 8k context.