What can the RTX 6000 Ada run?
| Text models: 184 | Visual models: 12 | Strongest first |
- Memory
- 48 GB
- Bandwidth
- 960 GB/s
- Maker
- NVIDIA
Open the calculator on this machineEvery model for 48 GB of VRAM
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Qwen3.8 27B | 8-bit | 256K | 23 tok/s | Chat, Coding, Agents, Docs |
| Muse Glimmer 30B | 8-bit | 128K | 21 tok/s | Docs |
| Qwen3.5 122B A10B | 2-bit | 32K | 115 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Super 120B A12B | 2-bit | 184K | 82 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3.5 Bolt 74B A9B | 4-bit | 256K | 73 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.6 35B A3B | 8-bit | 256K | 138 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 31B | 8-bit | 256K | 20 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 26B A4B | 8-bit | 256K | 128 tok/s | Chat, Coding, Agents, Docs |
| GLM-4.7 Flash | 8-bit | 198K | 132 tok/s | Chat, Coding, Agents |
| Gemma 4 12B | 8-bit | 256K | 49 tok/s | Chat, Coding, Docs |
| Mistral Small 4 119B | 2-bit | 136K | 173 tok/s | Chat, Docs |
| gpt-oss 120B | 2-bit | 104K | 200 tok/s | Chat, Coding |
| Qwen3 Coder Next 80B | 3-bit | 256K | 193 tok/s | Coding, Agents |
| Nemotron 3.5 Lightning 30B A3B | 8-bit | 256K | 126 tok/s | Agents, Docs |
| Devstral Small 2 24B | FP8 | 128K | 26 tok/s | Coding |
| LFM2.5 24B A2B | 8-bit | 128K | 180 tok/s | Agents |
| Mistral Small 3.2 24B | 8-bit | 128K | 26 tok/s | Chat, Coding, Agents |
| Ministral 3 14B | 8-bit | 192K | 45 tok/s | Coding, Agents, Docs |
| gpt-oss 20B | MXFP4 | 128K | 190 tok/s | Agents |
| Ministral 3 8B | 8-bit | 256K | 70 tok/s | Docs |
| Llama 3.3 70B | 3-bit | 36K | 19 tok/s | Chat, Coding, Agents, Docs |
| Olmo 3.1 32B | 8-bit | 64K | 19 tok/s | Chat, Coding, Agents, Docs |
| Llama 3.1 8B | 8-bit | 128K | 78 tok/s | Chat, Coding, Agents, Docs |
| Qwen3 14B | 8-bit | 40K | 43 tok/s | Not rated |
| Qwen3 VL 8B Instruct | 8-bit | 256K | 72 tok/s | Text and images |
| Qwen3 8B | 8-bit | 40K | 77 tok/s | Not rated |
| Qwen3 32B | 8-bit | 40K | 19 tok/s | Not rated |
| Qwen3.6 27B | 8-bit | 256K | 23 tok/s | Text and images |
| Qwen2.5 Coder 14B Instruct | 8-bit | 32K | 42 tok/s | Coding |
| Qwen3 Coder 30B A3B Instruct | 8-bit | 152K | 131 tok/s | Coding |
| Qwen3.5 27B | 8-bit | 256K | 23 tok/s | Text and images |
| Qwen3.5 35B A3B | 8-bit | 256K | 138 tok/s | Text and images |
| NVIDIA Nemotron 3 Nano 30B A3B BF16 | 8-bit | 256K | 126 tok/s | Not rated |
| GLM 4.5 Air | 2-bit | 32K | 90 tok/s | Not rated |
| Qwen3 30B A3B | 8-bit | 40K | 131 tok/s | Not rated |
| Qwen3 30B A3B Instruct 2507 | 8-bit | 152K | 131 tok/s | Not rated |
| Laguna XS.2 | 8-bit | 256K | 139 tok/s | Not rated |
| OTel 2.0 LLM 31B IT | 8-bit | 256K | 20 tok/s | Not rated |
| Qwen2.5 32B Instruct | 8-bit | 32K | 19 tok/s | Not rated |
| MiniCPM V 4.7 35B A3B | 8-bit | 256K | 138 tok/s | Text and images |
| Qwen2.5 14B Instruct | 8-bit | 32K | 42 tok/s | Not rated |
| Laguna XS 2.1 | 8-bit | 256K | 139 tok/s | Not rated |
| Diffusiongemma 26B A4B IT | 8-bit | 256K | 123 tok/s | Text and images |
| Qwen2.5 VL 32B Instruct | 8-bit | 44K | 19 tok/s | Text and images |
| Qwen2.5 Coder 32B Instruct | 8-bit | 32K | 19 tok/s | Coding |
| Laguna S 2.1 | 2-bit | 72K | 132 tok/s | Not rated |
| Qwen3 VL 30B A3B Instruct | 8-bit | 144K | 131 tok/s | Text and images |
| EXAONE 3.5 7.8B Instruct | 8-bit | 32K | 79 tok/s | Not rated |
| DeepSeek R1 0528 Qwen3 8B | 8-bit | 128K | 77 tok/s | Not rated |
| Gemma 3 12B IT | 8-bit | 128K | 52 tok/s | Text and images |
| Phi 3.5 Mini Instruct | 8-bit | 104K | 146 tok/s | Not rated |
| DeepSeek R1 Distill Qwen 32B | 8-bit | 48K | 19 tok/s | Not rated |
| Gemma 3 27B IT | 8-bit | 128K | 23 tok/s | Text and images |
| Ornith 1.5 35B A3B | 8-bit | 256K | 138 tok/s | Not rated |
| Ornith 1.5 122B | 2-bit | 32K | 98 tok/s | Not rated |
| GPT Neox 20B | 8-bit | 2K | 30 tok/s | Not rated |
| Apertus 8B Instruct 2509 | 8-bit | 64K | 78 tok/s | Not rated |
| Phi 4 | 8-bit | 16K | 42 tok/s | Not rated |
| Phi 3 Mini 128K Instruct | 8-bit | 104K | 146 tok/s | Not rated |
| Qwen3 VL 32B Instruct | 8-bit | 44K | 19 tok/s | Text and images |
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 184 are shown.
Image and video models
| Model | Best fit | Per output | Strongest at |
|---|---|---|---|
| MiniMax H3 | 16-bit | 2.9 min | Video |
| LTX-2.5 | 16-bit | 36 s | Video |
| FLUX.1 [dev] | 16-bit | 5.9 s | Images |
| Qwen-Image | 16-bit | 1.6 min | Images |
| FLUX.2 [dev] | 8-bit | 43 s | Images |
| Wan2.2 I2V A14B | 16-bit | 14 min | Video |
| Wan2.1 I2V 14B | 16-bit | 14 min | Video |
| Wan 2.1 T2V-14B | 16-bit | 18 min | Video |
| Krea 2 Turbo | 16-bit | 5.9 s | Images |
| Qwen Image 2512 | 16-bit | 1.6 min | Images |
| Wan 2.2 A14B | 16-bit | 14 min | Video |
| HunyuanVideo 1.5 | 16-bit | 17 min | Video |
Only just too big
These need up to a quarter more memory than the RTX 6000 Ada has. Spilling into system RAM or a lower precision can get them running, slowly.
Questions
What is the best LLM for the RTX 6000 Ada?
Of the open models we track, the strongest that run well on the RTX 6000 Ada are Qwen3.8 27B, Muse Glimmer 30B, Qwen3.5 122B A10B. 184 text models fit it in all.
Calculated from published specs, not benchmarked. All machines