Back to search results
gemma-4-31B-it · Q5_K_M
NVIDIA RTX 4080 Super16GBllama.cpp b4200L0 Self-reported
This page aggregates 1 real-world runs of gemma-4-31B-it (Q5_K_M) on NVIDIA RTX 4080 Super with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
25 tok/s
Decode
Decode speed
187.5 tok/s
Prefill
Prefill speed
0.53 s
TTFT
Time to first token
16 GB
VRAM
VRAM usage
L0 Self-reported
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- vLLM · IQ2_XS Decode 59.6 · Prefill 447 ·
- vLLM · Q5_K_M Decode 27.2 · Prefill 204 ·
- vLLM · Q4_K_M Decode 33.3 · Prefill 249.7 ·
- llama.cpp · IQ2_XS Decode 54.7 · Prefill 410.3 ·
- llama.cpp · Q5_K_M (current)Decode 25 · Prefill 187.5 ·
- llama.cpp · Q4_K_M Decode 30.5 · Prefill 228.8 ·
Core figures
- Decode (avg)
- 25 tok/s
- Prefill (avg)
- 187.5 tok/s
- TTFT (avg)
- 0.53 s
- VRAM (avg)
- 16 GB
- MTP acceptance rate
- —
- TTFB
- — GB
- Power draw
- — W
Configuration
Member-level fields are taken from the most recent run
- Model
- gemma-4-31B-it
- Quantization
- Q5_K_M
- Framework
- llama.cpp
- Version
- b4200
- Context length
- 4096 tokens
- Batch size
- 512
- 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
- NVIDIA RTX 4080 Super
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 16 GB
- OS
- Linux
- Driver
- 570.65
- CUDA
- 12.6
Sources & evidence
1 measured records in total, each traceable to its original source
b4200 · Linux · CUDA 12.6 · 4096 ctx
25 tok/s
Decode
187.5 tok/s
Prefill
0.53 s
TTFT
16 GB
VRAM
—
MTP
— W
Power draw
L0 理论估算（add-1000-records-batch3）