Back to search results
Qwen3.8-9B · IQ2_M
Apple M3 Max36GBllama.cpp v0.36L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.8-9B (IQ2_M) on Apple M3 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
136.47 tok/s
Decode
Decode speed
438.69 tok/s
Prefill
Prefill speed
1.95 s
TTFT
Time to first token
11.93 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- llama.cpp · IQ2_M (current)Decode 136.47 · Prefill 438.69 ·
- llama.cpp · IQ3_S Decode 126.59 · Prefill 664.89 ·
- Strata · Q2_K Decode 180.32 · Prefill 900.55 ·
Core figures
- Decode (avg)
- 136.47 tok/s
- Prefill (avg)
- 438.69 tok/s
- TTFT (avg)
- 1.95 s
- VRAM (avg)
- 11.93 GB
- MTP acceptance rate
- —
- TTFB
- 1.09 GB
- Power draw
- 56.5 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.8-9B
- Quantization
- IQ2_M
- Framework
- llama.cpp
- Version
- v0.36
- 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
- Apple M3 Max
- Nominal VRAM
- 36 GB
- Measured VRAM (avg)
- 11.93 GB
- OS
- macOS 15.0
- Driver
- Apple M3 Max
- CUDA
- N/A
- Power draw
- 56.5 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.36 · macOS 15.0 · CUDA N/A · 4096 ctx
136.47 tok/s
Decode
438.69 tok/s
Prefill
1.95 s
TTFT
11.93 GB
VRAM
—
MTP
56.5 W
Power draw
L2 batch seed 2026-10-10: llama.cpp on hw_id=4 running model_id=52 at IQ2_M