Back to search results
Qwen2.5-0.5B-Instruct · Q5_K_M
AMD Radeon AI PRO R970032GBStrata v0.50L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-0.5B-Instruct (Q5_K_M) on AMD Radeon AI PRO R9700 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
384.75 tok/s
Decode
Decode speed
2,096.67 tok/s
Prefill
Prefill speed
2.32 s
TTFT
Time to first token
3.08 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_XS Decode 346.96 · Prefill 1652.99 ·
- vLLM · Q5_K_S Decode 310.96 · Prefill 1151.48 ·
- Strata · Q3_K_M Decode 294.81 · Prefill 1272.19 ·
- Strata · Q5_K_M (current)Decode 384.75 · Prefill 2096.67 ·
Core figures
- Decode (avg)
- 384.75 tok/s
- Prefill (avg)
- 2,096.67 tok/s
- TTFT (avg)
- 2.32 s
- VRAM (avg)
- 3.08 GB
- MTP acceptance rate
- 0.78
- TTFB
- 3.08 GB
- Power draw
- 259.8 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-0.5B-Instruct
- Quantization
- Q5_K_M
- Framework
- Strata
- Version
- v0.50
- Context length
- 2048 tokens
- Batch size
- 512
- GPU layers
- 15
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Radeon AI PRO R9700
- Nominal VRAM
- 32 GB
- Measured VRAM (avg)
- 3.08 GB
- OS
- Windows 11 23H2
- Driver
- amdgpu-pro 24.20
- CUDA
- N/A
- Power draw
- 259.8 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.50 · Windows 11 23H2 · CUDA N/A · 2048 ctx
384.75 tok/s
Decode
2,096.67 tok/s
Prefill
2.32 s
TTFT
3.08 GB
VRAM
0.78
MTP
259.8 W
Power draw
L2 batch seed 2026-10-13: Strata on hw_id=79 running model_id=59 at Q5_K_M