Back to search results
Qwen3-4B-Thinking-2507 · Q5_K_S
AMD Radeon RX 9070 XT16GBvLLM v0.36L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (Q5_K_S) on AMD Radeon RX 9070 XT with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
103.28 tok/s
Decode
Decode speed
532.39 tok/s
Prefill
Prefill speed
0.88 s
TTFT
Time to first token
9.66 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- vLLM · IQ2_XS Decode 186.6 · Prefill 815.85 ·
- vLLM · Q5_K_S (current)Decode 103.28 · Prefill 532.39 ·
- Ollama · Q3_K_M Decode 127.72 · Prefill 433.24 ·
- Ollama · Q3_K_S Decode 126.74 · Prefill 598.67 ·
Core figures
- Decode (avg)
- 103.28 tok/s
- Prefill (avg)
- 532.39 tok/s
- TTFT (avg)
- 0.88 s
- VRAM (avg)
- 9.66 GB
- MTP acceptance rate
- —
- TTFB
- 0.83 GB
- Power draw
- 258.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- Q5_K_S
- Framework
- vLLM
- Version
- v0.36
- Context length
- 32768 tokens
- Batch size
- 8
- GPU layers
- 999
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Radeon RX 9070 XT
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 9.66 GB
- OS
- Windows 11 23H2
- Driver
- amdgpu 6.11
- CUDA
- N/A
- Power draw
- 258.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.36 · Windows 11 23H2 · CUDA N/A · 32768 ctx
103.28 tok/s
Decode
532.39 tok/s
Prefill
0.88 s
TTFT
9.66 GB
VRAM
—
MTP
258.3 W
Power draw
L2 batch seed 2026-10-13: vLLM on hw_id=16 running model_id=57 at Q5_K_S