Can I run this LLM?
Can a NVIDIA RTX Pro 6000 Blackwell (96GB) run Mellum2.1 12B A2.5B Thinking (MoE)?
Mellum · Apache 2.0 · estimated for a 96.0 GB machine at 4K context.
Yes — it runs great.
Runs greatBest fit: Mellum2.1 12B A2.5B Thinking (MoE) at FP16 · 24.3 GB weights · 152–228 tok/s on a NVIDIA RTX Pro 6000 Blackwell (96GB).
Pack & allocation notes: KV estimates assume a 1,024-token sliding window on 21 of 28 layers; actual runtime allocation may be higher. MXFP4_MOE is not listed. The FP16 label uses BF16 weights.
Arena Text Overall · style-controlled
UnratedNo exact matchNo reviewed exact Arena match
No reviewed exact model/configuration match in the Arena Text Overall style-controlled snapshot. No family or editorial score is substituted. Local quantization quality is unmeasured.
Snapshot published 2026-10-08 · snapshot imported 2026-10-11 UTC · validated last-good data · manual refresh
Arena leaderboard dataset · Arena / lmarena-ai, Arena Leaderboard Dataset (CC BY 4.0). Selected Text Overall style-controlled rows; field names adapted, ratings unchanged. · dataset CC BY 4.0 · revision e16f90232986
- Total memory at 4K context
- 25.2 GB
- Weights + KV cache + runtime overhead
- Estimated generation
- 152–228 tok/s
- Estimated first token
- ~0.3s
Every quantization on this machine
On small screens, each quant lists every metric below its name. On wider screens, scroll sideways for total memory, speed and fit. Keyboard: focus the table, then use the arrow keys. Quant names stay visible.
| Quant | Weights | Total memory | Speed | Fit |
|---|---|---|---|---|
| Q4_K_M | 8.1 GB | 9.0 GB | 246–370 tok/s | Runs great |
| Q6_K | 10.9 GB | 11.8 GB | 222–334 tok/s | Runs great |
| Q8_0 | 12.9 GB | 13.8 GB | 208–312 tok/s | Runs great |
| FP16 | 24.3 GB | 25.2 GB | 152–228 tok/s | Runs great |
Ran Mellum2.1 12B A2.5B Thinking (MoE) on a NVIDIA RTX Pro 6000 Blackwell (96GB)? Add your real numbers →