Can Qwen2.5 VL 3B Instruct run on the L40S?
Yes. At 8K context the best fit is F16, using about 6.7 GB of the 45.6 GB usable on the L40S. Estimated speed ceiling: ~140 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|
| 2-bit (Q2_K) • | 2.0 GB Fits | 2.2 GB Fits | 3.0 GB Fits | 678 |
| 3-bit (Q3_K_M) • | 2.3 GB Fits | 2.5 GB Fits | 3.3 GB Fits | 543 |
| 4-bit (Q4_K_M) • | 2.6 GB Fits | 2.8 GB Fits | 3.6 GB Fits | 448 |
| 5-bit (Q5_K_M) • | 2.9 GB Fits | 3.0 GB Fits | 3.9 GB Fits | 388 |
| 6-bit (Q6_K) • | 3.2 GB Fits | 3.3 GB Fits | 4.2 GB Fits | 340 |
| 8-bit (Q8_0) • | 3.9 GB Fits | 4.0 GB Fits | 4.9 GB Fits | 263 |
| 16-bit (F16/BF16) • | 6.6 GB Fits | 6.7 GB Fits | 7.6 GB Fits | 140 |
• real GGUF file size from unsloth/Qwen2.5-VL-3B-Instruct-GGUF. Speed is a bandwidth ceiling (864 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs Qwen2.5 VL 3B Instruct: Arc B580 (12 GB, launch MSRP $249) at F16, fully in VRAM at 8K context.
- Largest popular model that fits the L40S: Qwen3.6 35B A3B (36B, MoE) at Q8_0, about 36.2 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
Qwen2.5 VL 3B Instruct on every GPU and Mac · Every model that runs on the L40S · Other models that fit 48 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (Qwen/Qwen2.5-VL-3B-Instruct), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.