Back to search results
Llama-3.2-3B-Instruct · Q4_K_M
NVIDIA RTX 408016GBStrata v0.32L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (Q4_K_M) on NVIDIA RTX 4080 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
341.28 tok/s
Decode
Decode speed
1,116.7 tok/s
Prefill
Prefill speed
0.5 s
TTFT
Time to first token
7.27 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_S Decode 351.55 · Prefill 1651.53 ·
- llama.cpp · Q5_K_S Decode 338.44 · Prefill 1522.16 ·
- Strata · Q4_K_M (current)Decode 341.28 · Prefill 1116.7 ·
Core figures
- Decode (avg)
- 341.28 tok/s
- Prefill (avg)
- 1,116.7 tok/s
- TTFT (avg)
- 0.5 s
- VRAM (avg)
- 7.27 GB
- MTP acceptance rate
- 0.65
- TTFB
- 2.73 GB
- Power draw
- 273.2 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.2-3B-Instruct
- Quantization
- Q4_K_M
- Framework
- Strata
- Version
- v0.32
- Context length
- 32768 tokens
- Batch size
- 1024
- GPU layers
- 5
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 4080
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 7.27 GB
- OS
- Windows 11 24H2
- Driver
- 570.86
- CUDA
- 13.0
- Power draw
- 273.2 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.32 · Windows 11 24H2 · CUDA 13.0 · 32768 ctx
341.28 tok/s
Decode
1,116.7 tok/s
Prefill
0.5 s
TTFT
7.27 GB
VRAM
0.65
MTP
273.2 W
Power draw
L2 batch seed 2026-10-18: Strata on hw_id=41 running model_id=70 at Q4_K_M