Back to search results
Qwen3-Embedding-0.6B · IQ3_S
AMD Radeon RX 9070 XT16GBllama.cpp v0.29L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-Embedding-0.6B (IQ3_S) on AMD Radeon RX 9070 XT with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
308.65 tok/s
Decode
Decode speed
1,242.43 tok/s
Prefill
Prefill speed
2.31 s
TTFT
Time to first token
2.93 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- llama.cpp · IQ3_S (current)Decode 308.65 · Prefill 1242.43 ·
- llama.cpp · Q4_K_S Decode 280.71 · Prefill 1416.08 ·
- vLLM · IQ1_S Decode 320.75 · Prefill 1355.61 ·
- vLLM · Q5_K_M Decode 254.02 · Prefill 1313.89 ·
Core figures
- Decode (avg)
- 308.65 tok/s
- Prefill (avg)
- 1,242.43 tok/s
- TTFT (avg)
- 2.31 s
- VRAM (avg)
- 2.93 GB
- MTP acceptance rate
- —
- TTFB
- 2.47 GB
- Power draw
- 262 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-Embedding-0.6B
- Quantization
- IQ3_S
- Framework
- llama.cpp
- Version
- v0.29
- Context length
- 4096 tokens
- Batch size
- 256
- GPU layers
- 99
- 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 XT
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 2.93 GB
- OS
- Ubuntu 24.04
- Driver
- amdgpu 6.10
- CUDA
- N/A
- Power draw
- 262 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.29 · Ubuntu 24.04 · CUDA N/A · 4096 ctx
308.65 tok/s
Decode
1,242.43 tok/s
Prefill
2.31 s
TTFT
2.93 GB
VRAM
—
MTP
262 W
Power draw
L2 batch seed 2026-10-13: llama.cpp on hw_id=16 running model_id=62 at IQ3_S