Qwen2.5 VL 32B Instruct on H100 SXM: VRAM, speed and fit
Qwen2.5 VL 32B Instruct at 8-bit runs on the H100 SXM with 125K of context at about 60 tokens a second.
Text generation · Chat · Ollama / LM Studio
Runs comfortably · on H100 SXM · 8-bit (Q8_0) · Ollama / LM Studio · Chat
- GPU memory
- 65.2 GB
- of 80.0 GB
- Context
- 125K
- auto
- Speed
- 60 tok/s
- 32 when full
- First reply
- 0.4 s
- 2K prompt
Where the memory goes
- Model
- 33.1 GB
- Context
- 31.3 GB
- Overhead
- 918 MB
- OS
- 0.0 GB
- Free
- 14.8 GB
- Memory
- 65.2 GB of 80.0 GB
- Context
- 125K tokens · auto
- Speed
- 60 tok/s · 32 when full
- First reply
- 0.4 s for a 2K prompt
- Running cost
- $11/mo at home · $122/mo on a cloud 4× RTX A4500
Validation: Validated
| Check | Target | Result | Status |
|---|---|---|---|
| Fits in GPU memory | 80.0 GB usable | 65.2 GB needed | Pass |
| Headroom | 5% free | 14.8 GB free (18%) | Pass |
| Context for chat | 16K or more | 125K | Pass |
| Writing speed | 15 tok/s or more | 60 tok/s | Pass |
| First reply | under 3 s, 2K prompt | 0.4 s | Pass |
Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.
The setup
- Model
- Qwen2.5 VL 32B Instruct
- Parameters
- 33.5B
- Precision
- 8-bit (Q8_0)
- Context
- 125K, auto
- Chats at once
- 1
- Use
- Chat
- Runs in
- Ollama / LM Studio
- Hardware
- H100 SXM
- GPU can use
- 80.0 GB of 80.0 GB
Running cost
- At home
- $11/mo
- 500 W at the wall
- Renting instead
- $122/mo
- 4× RTX A4500 at $1.00/h
4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 22, 2026.
How to run it
- llama.cpp
llama-server -hf bartowski/Qwen_Qwen2.5-VL-32B-Instruct-GGUF- vLLM
vllm serve Qwen/Qwen2.5-VL-32B-Instruct
Qwen2.5 VL 32B Instruct on similar machines
- 2× RTX PRO 6000 BlackwellRuns
- 4× RTX PRO 6000 BlackwellRuns
- 8× RTX PRO 6000 BlackwellRuns
- A100 80 GB SXMRuns
- 4× H200 NVLRuns
- 8× H200 NVLRuns
All machines: Qwen2.5 VL 32B Instruct VRAM requirements. Everything that runs here: what can the H100 SXM run?
Other models that run on H100 SXM
Questions
Can Qwen2.5 VL 32B Instruct run on H100 SXM?
Qwen2.5 VL 32B Instruct at 8-bit runs on the H100 SXM with 125K of context at about 60 tokens a second.
Report VR-16N47ZT, calculated Nov 22, 2026 from published specs, not benchmarked.