Benchmark search

Try these searches

Found 18 run configurations

Each row is a run configuration (model + hardware + framework + quantization); repeated runs of the same setup are merged — raw counts appear in each result’s source line.

ModelMiniMax-H3

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 3090 · 24GB · Diffusers

Other

2026-10-06

7.1 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 3090 · 24GB · Diffusers

Other

2026-10-06

14.2 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 3090 · 24GB · vLLM-Omni

Other

2026-10-06

7.1 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 3090 · 24GB · vLLM-Omni

Other

2026-10-06

14.2 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 5070 Ti · 16GB · Diffusers

Other

2026-10-06

7.1 s

Gen time

16 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 Diffusers 0.32.0 1080p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 5070 Ti · 16GB · Diffusers

Other

2026-10-06

14.2 s

Gen time

16 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 5070 Ti · 16GB · vLLM-Omni

Other

2026-10-06

7.1 s

Gen time

16 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 5070 Ti · 16GB · vLLM-Omni

Other

2026-10-06

14.2 s

Gen time

16 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 5090 · 32GB · Diffusers

Other

2026-10-06

7.1 s

Gen time

32 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 5090 · 32GB · Diffusers

Other

2026-10-06

14.2 s

Gen time

32 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 5090 · 32GB · vLLM-Omni

Other

2026-10-06

7.1 s

Gen time

32 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 vLLM-Omni 0.7.3 1080p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 5090 · 32GB · vLLM-Omni

Other

2026-10-06

14.2 s

Gen time

32 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 4090 · 24GB · Diffusers

Other

2026-10-06

7.1 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 4090 · 24GB · Diffusers

Other

2026-10-06

14.2 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 Diffusers 0.32.0 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · FP8

L0 Self-reported

NVIDIA RTX 4090 · 24GB · vLLM-Omni

Other

2026-10-06

7.1 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 FP8 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX 4090 · 24GB · vLLM-Omni

Other

2026-10-06

14.2 s

Gen time

24 GB

VRAM

Config (latest run)
Linux CUDA 12.6 BF16 vLLM-Omni 0.7.3 720p · 5s

Sources: Other · 1 runs · 1 independent sources

MiniMax-H3 · Online dynamic FP8

L0 Self-reported

NVIDIA DGX Spark · 128GB · vLLM-Omni

GitHub

2026-09-27

80.58 s

Gen time

89.17 GB

VRAM

Config (latest run)
Linux ARM64 CUDA 13.0 Online dynamic FP8 vLLM-Omni vLLM-Omni 0.1.dev2381+g310b4b477 768x448 · 56 frames · 20 steps · 24fps

Sources: GitHub · 1 runs · 1 independent sources

MiniMax-H3 · BF16

L0 Self-reported

NVIDIA RTX PRO 6000 Blackwell · 96GB · Diffusers

Hugging Face

2026-09-27

317 s

Gen time

78.54 GB

VRAM

Config (latest run)
Linux BF16 Diffusers Diffusers PR14371 665f5782 1344x768 · 124 frames · 30 steps · 24fps

Sources: Hugging Face · 1 runs · 1 independent sources

18 results

Related models
Related hardware
Related frameworks
Related quantizations