Back to search results
Qwen3-4B-Thinking-2507 · IQ3_S
NVIDIA RTX 4060 Ti 16GB16GBvLLM v0.27L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (IQ3_S) on NVIDIA RTX 4060 Ti 16GB with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
108.19 tok/s
Decode
Decode speed
350.2 tok/s
Prefill
Prefill speed
2.25 s
TTFT
Time to first token
7.64 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- vLLM · IQ3_S (current)Decode 108.19 · Prefill 350.2 ·
- vLLM · Q3_K_M Decode 119.54 · Prefill 394.34 ·
- Strata · IQ3_XS Decode 160.63 · Prefill 538.1 ·
- Strata · Q3_K_L Decode 140.78 · Prefill 483.92 ·
Core figures
- Decode (avg)
- 108.19 tok/s
- Prefill (avg)
- 350.2 tok/s
- TTFT (avg)
- 2.25 s
- VRAM (avg)
- 7.64 GB
- MTP acceptance rate
- —
- TTFB
- 0.87 GB
- Power draw
- 274 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- IQ3_S
- Framework
- vLLM
- Version
- v0.27
- Context length
- 32768 tokens
- Batch size
- 16
- 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 4060 Ti 16GB
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 7.64 GB
- OS
- macOS 15.0
- Driver
- 555.85
- CUDA
- 12.4
- Power draw
- 274 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.27 · macOS 15.0 · CUDA 12.4 · 32768 ctx
108.19 tok/s
Decode
350.2 tok/s
Prefill
2.25 s
TTFT
7.64 GB
VRAM
—
MTP
274 W
Power draw
L2 batch seed 2026-10-13: vLLM on hw_id=3 running model_id=57 at IQ3_S