Back to search results
Qwen3.5-4B · Q3_K_M
NVIDIA RTX 509032GBvLLM v0.25L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-4B (Q3_K_M) on NVIDIA RTX 5090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
356.1 tok/s
Decode
Decode speed
1,633.97 tok/s
Prefill
Prefill speed
2.15 s
TTFT
Time to first token
7.81 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- llama.cpp · Q3_K_L Decode 338.21 · Prefill 1409.78 ·
- vLLM · Q3_K_M (current)Decode 356.1 · Prefill 1633.97 ·
- vLLM · Q3_K_S Decode 324.89 · Prefill 1253.12 ·
- Strata · Q4_K_S Decode 353.67 · Prefill 1676.75 ·
Core figures
- Decode (avg)
- 356.1 tok/s
- Prefill (avg)
- 1,633.97 tok/s
- TTFT (avg)
- 2.15 s
- VRAM (avg)
- 7.81 GB
- MTP acceptance rate
- —
- TTFB
- 2.85 GB
- Power draw
- 487.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-4B
- Quantization
- Q3_K_M
- Framework
- vLLM
- Version
- v0.25
- Context length
- 8192 tokens
- Batch size
- 32
- GPU layers
- 999
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 5090
- Nominal VRAM
- 32 GB
- Measured VRAM (avg)
- 7.81 GB
- OS
- Windows 11 23H2
- Driver
- 560.94
- CUDA
- 12.6
- Power draw
- 487.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.25 · Windows 11 23H2 · CUDA 12.6 · 8192 ctx
356.1 tok/s
Decode
1,633.97 tok/s
Prefill
2.15 s
TTFT
7.81 GB
VRAM
—
MTP
487.3 W
Power draw
L2 batch seed 2026-10-10: vLLM on hw_id=13 running model_id=48 at Q3_K_M