What can the DGX Station GB300 run?
| Text models: 83 | Visual models: 0 | Strongest first |
- Memory
- 252 GB
- Bandwidth
- 7,100 GB/s
- Maker
- NVIDIA
Open the calculator on this machineDGX Station GB300 in our inventory
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Kimi K3 | 1.6-bit | 1M | 20 tok/s | Chat, Coding, Agents, Docs |
| MiMo V2.6 Pro | 1.6-bit | 832K | 168 tok/s | Chat, Coding, Agents, Docs |
| DeepSeek V4.1 Flash | 1.6-bit | 1M | 322 tok/s | Chat, Coding, Agents, Docs |
| GLM-5.3 | 1.6-bit | 960K | 158 tok/s | Chat, Coding, Agents, Docs |
| GLM-5.3 Flash | 5-bit | 1M | 188 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.8 2.4T A95B | 1.6-bit | 256K | 22 tok/s | Chat, Coding, Agents |
| Qwen3.8 Flash-Next | 8-bit | 256K | 239 tok/s | Chat, Coding, Agents |
| Muse Spark Open 816B A42B | 1.6-bit | 1M | 183 tok/s | Chat, Docs |
| MiMo V2.6 Flash | FP4 + FP8 | 1M | 198 tok/s | Coding, Agents |
| MiniMax M3 | 3-bit | 352K | 171 tok/s | Docs |
| DeepSeek V4 Flash | FP4 + FP8 | 1M | 157 tok/s | Docs |
| DeepSeek V4 Pro | 2-bit | 1M | 29 tok/s | Chat, Coding, Agents, Docs |
| Kimi K2.6 | 1.6-bit | 256K | 200 tok/s | Chat, Coding, Agents, Docs |
| Hy3 | 6-bit | 40K | 121 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Ultra 550B A55B | 2-bit | 256K | 93 tok/s | Chat, Coding, Docs |
| Qwen3.5 397B A17B | 4-bit | 256K | 175 tok/s | Agents, Docs |
| MiniMax M2.7 | FP8 | 96K | 174 tok/s | Coding, Agents |
| DeepSeek V3.2 | 2-bit | 52K | 155 tok/s | Chat, Coding, Agents, Docs |
| Mistral Medium 3.5 128B | FP8 | 256K | 34 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.5 122B A10B | 8-bit | 256K | 198 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Super 120B A12B | 8-bit | 256K | 127 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3.5 Bolt 74B A9B | 8-bit | 256K | 138 tok/s | Chat, Coding, Agents, Docs |
| Mistral Small 4 119B | FP8 | 256K | 287 tok/s | Chat, Docs |
| Qwen3 Coder Next 80B | 8-bit | 256K | 283 tok/s | Coding, Agents |
| Llama 3.3 70B | 8-bit | 128K | 58 tok/s | Chat, Coding, Agents, Docs |
| Inkling | 1.6-bit | 4K | 177 tok/s | Text and images |
| DeepSeek R1 | 2-bit | 160K | 150 tok/s | Not rated |
| DeepSeek V3 | 2-bit | 160K | 150 tok/s | Not rated |
| GLM 5 | 1.6-bit | 198K | 155 tok/s | Not rated |
| GLM 4.5 Air | 8-bit | 128K | 168 tok/s | Not rated |
| Inkling Small | 6-bit | 4K | 227 tok/s | Text and images |
| MiniMax M2.5 | FP8 | 96K | 165 tok/s | Not rated |
| GLM 5.1 | 1.6-bit | 198K | 155 tok/s | Not rated |
| Laguna M.1 | 8-bit | 52K | 109 tok/s | Not rated |
| Laguna S 2.1 | 8-bit | 1M | 213 tok/s | Not rated |
| Qwen3 VL 235B A22B Instruct | 8-bit | 28K | 97 tok/s | Text and images |
| GLM 5.2 | 1.6-bit | 960K | 156 tok/s | Not rated |
| Qwen3 235B A22B | 8-bit | 32K | 97 tok/s | Not rated |
| Ornith 1.5 122B | 8-bit | 256K | 177 tok/s | Not rated |
| Kimi K2 Instruct | 1.6-bit | 128K | 186 tok/s | Not rated |
| Ornith 1.5 397B | 4-bit | 256K | 151 tok/s | Not rated |
| Ornith 1.0 397B | 4-bit | 256K | 172 tok/s | Not rated |
| Step 3.5 Flash | 8-bit | 104K | 212 tok/s | Not rated |
| DeepSeek V3 0324 | 2-bit | 160K | 150 tok/s | Not rated |
| Qwen2.5 72B Instruct | 8-bit | 32K | 57 tok/s | Not rated |
| Qwen3 Next 80B A3B Instruct | 8-bit | 256K | 283 tok/s | Not rated |
| DeepSeek R1 0528 | 2-bit | 160K | 150 tok/s | Not rated |
| GLM 4.5 | 4-bit | 96K | 101 tok/s | Not rated |
| Llama 4 Scout 17B 16E Instruct | 8-bit | 672K | 148 tok/s | Text and images |
| MiniMax M2 | FP8 | 96K | 165 tok/s | Not rated |
| MiMo V2.5 | 6-bit | 52K | 174 tok/s | Not rated |
| Qwen2.5 VL 72B Instruct | 8-bit | 125K | 56 tok/s | Text and images |
| Solar Open 100B | 8-bit | 128K | 183 tok/s | Not rated |
| GLM 5.3 Flash BF16 | 5-bit | 1M | 187 tok/s | Text and images |
| Qwen3 VL 235B A22B Thinking | 8-bit | 28K | 97 tok/s | Text and images |
| Qwen3 235B A22B Instruct 2507 | 8-bit | 32K | 97 tok/s | Not rated |
| NVIDIA Nemotron 3.5 Super VL 120B A12B BF16 | 8-bit | 1M | 127 tok/s | Text and images |
| Hy3 Preview | 6-bit | 32K | 121 tok/s | Not rated |
| Qwen3 Coder 480B A35B Instruct | 3-bit | 80K | 140 tok/s | Coding |
| GLM 4.7 | 4-bit | 96K | 101 tok/s | Not rated |
Best fit is the highest quality, up to 8-bit, that leaves the use case’s context. Ranked by our use-case scores, then by Hugging Face downloads. Speeds are rough, plus or minus 25%. The top 60 of 83 are shown.
Questions
What is the best LLM for the DGX Station GB300?
Of the open models we track, the strongest that run well on the DGX Station GB300 are Kimi K3, MiMo V2.6 Pro, DeepSeek V4.1 Flash. 83 text models fit it in all.
Calculated from published specs, not benchmarked. All machines