Can the RTX PRO 6000 Blackwell 96GB Run Qwen2.5 32B at Q6_K?
Runs
Yes. Qwen2.5 32B at Q6_K is a 26.9 GB download (Hugging Face file listing for bartowski/Qwen2.5-32B-Instruct-GGUF), and the RTX PRO 6000 Blackwell 96GB has 96 GB of VRAM, leaving about 69.1 GB for the KV cache and runtime. No published speed measurement for this pair passes SpecPicks' sanity checks yet, so no tokens-per-second figure is quoted.
Buy the RTX PRO 6000 Blackwell 96GB
No Amazon listing for this card is tracked today — see live eBay listings.
Published runs for Qwen2.5 32B on the RTX PRO 6000 Blackwell 96GB
No published run for this pair has passed the checks yet.
More about this pair
How much VRAM does Qwen2.5 32B need?
The Q6_K file is 26.9 GB. Add roughly 1.5 GB for the runtime and an ~8K-token context, so about 28.4 GB of VRAM holds it comfortably. Longer contexts grow the KV cache and need more.
What speed does Qwen2.5 32B reach on the RTX PRO 6000 Blackwell 96GB?
No published run of Qwen2.5 32B on the RTX PRO 6000 Blackwell 96GB has been cited yet. Generation reads every weight once per token, so the RTX PRO 6000 Blackwell 96GB's 1792 GB/s of memory bandwidth divided by the 26.89 GB file puts the theoretical ceiling near 67 tokens per second. That is an upper bound, not a speed estimate: kernel, launch and compute overhead keep real runs below it.
Check another pair
See the full matrix → · All RTX PRO 6000 Blackwell 96GB benchmarks →
Qwen2.5 32B on other GPUs
- RTX 4060 — Offload
- RTX 3060 Ti — Offload
- RTX 3060 12GB — Offload
- Intel Arc B580 — Offload
- RTX 4070 SUPER — Offload
- RTX 4060 Ti 16GB — Offload
- RTX 5060 Ti 16GB — Offload
- Intel Arc A770 16GB — Offload
- Radeon RX 9070 XT — Offload
- RTX 4070 Ti SUPER — Offload
- RTX 4080 — Offload
- RTX 5070 Ti — Offload
- RTX 5080 — Offload
- RTX 3090 — Offload
- Radeon RX 7900 XTX — Offload
- RTX 4090 — Offload
- RTX 5090 — Runs
- RTX A6000 48GB — Runs
Other models on the RTX PRO 6000 Blackwell 96GB
- Llama 3.2 1B — Runs, 244 tok/s (1 run)
- Llama 3.2 3B — Runs
- Mistral 7B — Runs
- Qwen2.5 7B — Runs
- Llama 3.1 8B — Runs, 138 tok/s (1 run)
- DeepSeek-R1-Distill-Llama 8B — Runs
- Qwen3 8B — Runs
- Gemma 2 9B — Runs
- Gemma 3 12B — Runs
- Qwen2.5 14B — Runs, 81.9 tok/s (1 run)
- DeepSeek-R1-Distill-Qwen 14B — Runs
- Qwen3 14B — Runs, 114 tok/s (1 run)
- Phi-4 14B — Runs
- gpt-oss-20b — Runs
- Gemma 3 27B — Runs
- Gemma 2 27B — Runs
- Qwen3 30B-A3B — Runs
- Qwen3 32B — Runs
- Qwen2.5 32B — Runs, 56 tok/s (1 run)
- DeepSeek-R1-Distill-Qwen 32B — Runs
- Qwen3 32B — Runs
- DeepSeek-R1-Distill-Qwen 32B — Runs
- Gemma 3 27B — Runs
- Llama 3.1 70B — Runs, 28.8 tok/s median
- Llama 3.3 70B — Runs, 30.9 tok/s median
- DeepSeek-R1-Distill-Llama 70B — Runs, 28.4 tok/s (1 run)
How the checks work
The fit is the Q6_K file size against the card's VRAM, with 1.5 GB kept for the runtime and an ~8K context. A published run counts only if the pair fits, the run is Q4-class, its source does not call it an estimate, it is under the bandwidth ceiling (memory bandwidth ÷ GB read per token: the whole file for a dense model, only the active experts' weights for a mixture-of-experts model such as gpt-oss-20b or Qwen3 30B-A3B, which is why those can post figures above bandwidth ÷ file size), it is not slower than the same card on a model twice its size, a single run is not slower than the same card on a larger dense model, and it is not one figure pasted across many cards. The whole matrix is downloadable as a CSV under CC BY 4.0.
As an Amazon Associate, SpecPicks earns from qualifying purchases. As an eBay Partner Network member, SpecPicks may earn a commission on eBay links.
This page is editorial synthesis based on publicly available information. No independent first-party benchmarking is reported.