Back to search results
Llama-3.1-8B-Instruct · Q3_K_M
NVIDIA RTX 4080 Super16GBllama.cpp v0.52L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (Q3_K_M) on NVIDIA RTX 4080 Super with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
201.4 tok/s
Decode
Decode speed
1,061.69 tok/s
Prefill
Prefill speed
1.81 s
TTFT
Time to first token
13.52 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
17 days ago
Last verified
Performance
- llama.cpp · Q3_K_M (current)Decode 201.4 · Prefill 1061.69 ·
- Strata · IQ2_XS Decode 307.75 · Prefill 1104.97 ·
- Strata · Q5_K_M Decode 179.72 · Prefill 584.63 ·
- vLLM · IQ2_M Decode 224.52 · Prefill 1147.46 ·
- vLLM · Q3_K_L Decode 177.01 · Prefill 908.47 ·
- Ollama · IQ2_M Decode 204.79 · Prefill 1109.54 ·
Core figures
- Decode (avg)
- 201.4 tok/s
- Prefill (avg)
- 1,061.69 tok/s
- TTFT (avg)
- 1.81 s
- VRAM (avg)
- 13.52 GB
- MTP acceptance rate
- —
- TTFB
- 1.61 GB
- Power draw
- 303.4 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.1-8B-Instruct
- Quantization
- Q3_K_M
- Framework
- llama.cpp
- Version
- v0.52
- Context length
- 2048 tokens
- Batch size
- 512
- 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
- NVIDIA RTX 4080 Super
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 13.52 GB
- OS
- Ubuntu 24.04
- Driver
- 560.94
- CUDA
- 12.8
- Power draw
- 303.4 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.52 · Ubuntu 24.04 · CUDA 12.8 · 2048 ctx
201.4 tok/s
Decode
1,061.69 tok/s
Prefill
1.81 s
TTFT
13.52 GB
VRAM
—
MTP
303.4 W
Power draw
L2 batch seed 2026-10-19: llama.cpp on hw_id=20 running model_id=68 at Q3_K_M