Back to search results
gemma-4-E4B-it · Q3_K_M
NVIDIA H2097GBvLLM v0.34L2 Cross-framework verified
This page aggregates 1 real-world runs of gemma-4-E4B-it (Q3_K_M) on NVIDIA H20 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
78.41 tok/s
Decode
Decode speed
259.38 tok/s
Prefill
Prefill speed
0.43 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
12 days ago
Last verified
Performance
- llama.cpp · IQ2_XS Decode 95.17 · Prefill 436.06 ·
- vLLM · Q3_K_M (current)Decode 78.41 · Prefill 259.38 ·
Core figures
- Decode (avg)
- 78.41 tok/s
- Prefill (avg)
- 259.38 tok/s
- TTFT (avg)
- 0.43 s
- VRAM (avg)
- 13.52 GB
- MTP acceptance rate
- —
- TTFB
- 0.63 GB
- Power draw
- 332.7 W
Configuration
Member-level fields are taken from the most recent run
- Model
- gemma-4-E4B-it
- Quantization
- Q3_K_M
- Framework
- vLLM
- Version
- v0.34
- Context length
- 4096 tokens
- Batch size
- 8
- GPU layers
- 999
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA H20
- Nominal VRAM
- 97 GB
- Measured VRAM (avg)
- 13.52 GB
- OS
- Ubuntu 22.04
- Driver
- 570.86
- CUDA
- 12.8
- Power draw
- 332.7 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.34 · Ubuntu 22.04 · CUDA 12.8 · 4096 ctx
78.41 tok/s
Decode
259.38 tok/s
Prefill
0.43 s
TTFT
13.52 GB
VRAM
—
MTP
332.7 W
Power draw
L2 batch seed 2026-10-24: vLLM on hw_id=56 running model_id=67 at Q3_K_M