Back to search results
Qwen3-8B · Q3_K_L
AMD Radeon RX 907016GBllama.cpp v0.30L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-8B (Q3_K_L) on AMD Radeon RX 9070 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
112.13 tok/s
Decode
Decode speed
574.57 tok/s
Prefill
Prefill speed
0.69 s
TTFT
Time to first token
14.2 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 (current)Decode 112.13 · Prefill 574.57 ·
- llama.cpp · Q5_K_M Decode 84.7 · Prefill 430.21 ·
- Strata · Q4_K_S Decode 150.11 · Prefill 526.56 ·
- Strata · Q3_K_L Decode 126.35 · Prefill 621.6 ·
- Strata · Q3_K_S Decode 136.8 · Prefill 667.03 ·
Core figures
- Decode (avg)
- 112.13 tok/s
- Prefill (avg)
- 574.57 tok/s
- TTFT (avg)
- 0.69 s
- VRAM (avg)
- 14.2 GB
- MTP acceptance rate
- —
- TTFB
- 0.9 GB
- Power draw
- 247.8 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-8B
- Quantization
- Q3_K_L
- Framework
- llama.cpp
- Version
- v0.30
- Context length
- 4096 tokens
- Batch size
- 256
- GPU layers
- 20
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Radeon RX 9070
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 14.2 GB
- OS
- Windows 11 23H2
- Driver
- amdgpu-pro 24.20
- CUDA
- 12.4
- Power draw
- 247.8 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.30 · Windows 11 23H2 · CUDA 12.4 · 4096 ctx
112.13 tok/s
Decode
574.57 tok/s
Prefill
0.69 s
TTFT
14.2 GB
VRAM
—
MTP
247.8 W
Power draw
L2 batch seed 2026-10-18: llama.cpp on hw_id=21 running model_id=53 at Q3_K_L