MiniMax-H3 · FP8
This page aggregates 1 real-world runs of MiniMax-H3 (FP8) on NVIDIA RTX 3090 with vLLM-Omni, contributed by 1 independent source platforms; metrics are averages of published measurements.
7.1 s
Gen time
Per image / per clip
24 GB
VRAM
VRAM usage
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
No published records for this metric under the same model + hardware yet
Core figures
- Gen time (avg)
- 7.1 s
- VRAM (avg)
- 24 GB
- Power draw
- — W
- Output
- 720p · 5s
Configuration
Member-level fields are taken from the most recent run
- Model
- MiniMax-H3
- Quantization
- FP8
- Framework
- vLLM-Omni
- Version
- 0.7.3
- Output spec
- 720p · 5s
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 3090
- Nominal VRAM
- 24 GB
- Measured VRAM (avg)
- 24 GB
- OS
- Linux
- Driver
- 570.65
- CUDA
- 12.6
Get started
Original model resources
Base model resources may not match this quantization. Use the verified deployment weights above when available.
Sources & evidence
1 measured records in total, each traceable to its original source
0.7.3 · Linux · CUDA 12.6
7.1 s
Gen time
24 GB
VRAM
— W
Power draw
L0 理论估算（add-500-records-batch2）