What can the 8× RTX PRO 6000 Blackwell run?
| Text models: 53 | Visual models: 0 | Strongest first |
- Memory
- 768 GB
- Bandwidth
- 1,792 GB/s each
- Maker
- NVIDIA
Open the calculator on this machineRTX PRO 6000 Blackwell in our inventory
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Kimi K3 | 1.6-bit | 1M | 43 tok/s | Chat, Coding, Agents, Docs |
| MiMo V2.6 Pro | FP4 + FP8 | 1M | 33 tok/s | Chat, Coding, Agents, Docs |
| DeepSeek V4.1 Flash | FP4 + FP8 | 1M | 63 tok/s | Chat, Coding, Agents, Docs |
| GLM-5.3 | FP8 | 208K | 27 tok/s | Chat, Coding, Agents, Docs |
| GLM-5.3 Flash | FP8 | 1M | 57 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.8 2.4T A95B | 1.6-bit | 256K | 46 tok/s | Chat, Coding, Agents |
| Muse Spark Open 816B A42B | 6-bit | 1M | 32 tok/s | Chat, Docs |
| MiniMax M3 | 8-bit | 1M | 43 tok/s | Docs |
| DeepSeek V4 Pro | 2-bit | 1M | 55 tok/s | Chat, Coding, Agents, Docs |
| Kimi K2.6 | INT4 | 256K | 25 tok/s | Chat, Coding, Agents, Docs |
| Hy3 | 8-bit | 256K | 44 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Ultra 550B A55B | 8-bit | 256K | 19 tok/s | Chat, Coding, Docs |
| Qwen3.5 397B A17B | 8-bit | 256K | 55 tok/s | Agents, Docs |
| MiniMax M2.7 | FP8 | 200K | 85 tok/s | Coding, Agents |
| DeepSeek V3.2 | FP8 | 160K | 29 tok/s | Chat, Coding, Agents, Docs |
| Inkling | 6-bit | 4K | 33 tok/s | Text and images |
| DeepSeek R1 | FP8 | 160K | 28 tok/s | Not rated |
| DeepSeek V3 | FP8 | 160K | 28 tok/s | Not rated |
| GLM 5 | 6-bit | 198K | 31 tok/s | Not rated |
| Inkling Small | 8-bit | 4K | 82 tok/s | Text and images |
| MiniMax M2.5 | FP8 | 192K | 77 tok/s | Not rated |
| GLM 5.1 | 6-bit | 198K | 31 tok/s | Not rated |
| Laguna M.1 | 8-bit | 256K | 42 tok/s | Not rated |
| Qwen3 VL 235B A22B Instruct | 8-bit | 256K | 41 tok/s | Text and images |
| GLM 5.2 | 6-bit | 1M | 31 tok/s | Not rated |
| Qwen3 235B A22B | 8-bit | 40K | 41 tok/s | Not rated |
| Kimi K2 Instruct | 5-bit | 128K | 37 tok/s | Not rated |
| Ornith 1.5 397B | 8-bit | 256K | 41 tok/s | Not rated |
| Ornith 1.0 397B | 8-bit | 256K | 53 tok/s | Not rated |
| Step 3.5 Flash | 8-bit | 256K | 94 tok/s | Not rated |
| DeepSeek V3 0324 | FP8 | 160K | 28 tok/s | Not rated |
| DeepSeek R1 0528 | FP8 | 160K | 28 tok/s | Not rated |
| GLM 4.5 | 8-bit | 128K | 28 tok/s | Not rated |
| MiniMax M2 | FP8 | 192K | 77 tok/s | Not rated |
| MiMo V2.5 | FP8 | 1M | 59 tok/s | Not rated |
| GLM 5.3 Flash BF16 | 8-bit | 1M | 55 tok/s | Text and images |
| Qwen3 VL 235B A22B Thinking | 8-bit | 256K | 41 tok/s | Text and images |
| Qwen3 235B A22B Instruct 2507 | 8-bit | 256K | 41 tok/s | Not rated |
| Hy3 Preview | 8-bit | 256K | 44 tok/s | Not rated |
| Qwen3 Coder 480B A35B Instruct | 8-bit | 256K | 30 tok/s | Coding |
| GLM 4.7 | 8-bit | 198K | 28 tok/s | Not rated |
| GLM 5.3 BF16 | 6-bit | 1M | 31 tok/s | Not rated |
| GLM 4.6 | 8-bit | 198K | 29 tok/s | Not rated |
| DeepSeek V2 Chat | 8-bit | 160K | 46 tok/s | Not rated |
| MiMo V2 Flash | FP8 | 256K | 62 tok/s | Not rated |
| Command A Plus 05 2026 Bf16 | 8-bit | 195K | 43 tok/s | Text and images |
| K EXAONE 2.0 750B A37B | 6-bit | 256K | 35 tok/s | Not rated |
| MiniMax M2.1 | FP8 | 192K | 77 tok/s | Not rated |
| Beam 1.0 501B A23B | 8-bit | 1M | 44 tok/s | Not rated |
| Hy4 Preview | 6-bit | 1M | 27 tok/s | Not rated |
| Intern S1 | 8-bit | 64K | 34 tok/s | Text and images |
| Step 3.7 Flash | 8-bit | 256K | 93 tok/s | Text and images |
| K2 Horizon 375B A23B | 8-bit | 512K | 43 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%.
Questions
What is the best LLM for the 8× RTX PRO 6000 Blackwell?
Of the open models we track, the strongest that run well on the 8× RTX PRO 6000 Blackwell are Kimi K3, MiMo V2.6 Pro, DeepSeek V4.1 Flash. 53 text models fit it in all.
Calculated from published specs, not benchmarked. All machines