MiMo VL 7B SFT 2508
Aug 2025
MiMo VL 7B SFT 2508Compare inon
MiMo VL 7B SFT 2508 measured on 2 of 15 Vision & Documents benchmarks2
BenchmarkByScoreRank
Legacy · saturated or no longer reported2
No benchmark matches
—per task?
self-host onlyprice
—output speed?
128Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGemini 3 Flash Previewunstated×3.1—$0.50 / $3
OpenQwen3 VL 8B Instructunstated+16%—$0.18 / $0.69BetterMiMo v2 Omniunstated×2.8—$0.14 / $0.28
BetterKimi K2.5unstatedOpen×2.6—$0.60 / $3Memory (4-bit): 4.9 GB → 616 GB
BetterGPT-5unstated×2.5—$1.25 / $10ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Gemini 3 Flash Previewunstated×3.1—$0.50 / $3MiMo v2 Omniunstated×2.8—$0.14 / $0.28Kimi K2.5unstatedOpen×2.6—$0.60 / $3Memory (4-bit): 4.9 GB → 616 GBGPT-5unstated×2.5—$1.25 / $10Gemma 4 31BunstatedOpen+78%—$0.10 / $0.34Memory (4-bit): 4.9 GB → 19 GBModelScoreCostPer taskWhat changes
Qwen3 VL 8B InstructunstatedOpen+16%—$0.18 / $0.69MiMo VL 7B RL 2508Open+11%——Qwen3 VL 4B InstructunstatedOpensame—$0.10 / $0.60OInternVL3 5 8B InstructOpensame——Context: 128K → 41K+1OInternVL3 5 4B InstructOpensame——Context: 128K → 41KModelScoreCostPer taskWhat changes
MiMo v2 Omniunstated×2.8—$0.14 / $0.28MiMo VL 7B RL 2508Open+11%——MiMo v2.6 FlashthinkingOpenn/a—$0.062Memory (4-bit): 4.9 GB → 186 GB+4MiMo v2.6 ProthinkingOpenn/a—$0.13Memory (4-bit): 4.9 GB → 615 GB+4MiMo v2.6 Pro UltraSpeedOpenn/a—$4.35 / $8.70Memory (4-bit): not verifiedMiMo v2.5thinkingOpenn/a—$0.019Memory (4-bit): 4.9 GB → 186 GBMiMo v2.5 ProthinkingOpenn/a—$0.054Memory (4-bit): 4.9 GB → 614 GB+1MiMo v2 Omni 0327n/a——All Xiaomi models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceMITfree for commercial use
Benchmarks
in Vision & DocumentsVideo-MME 2 · with subtitles/audio14.7#16
Video-MME 2 · without subtitles/audio10.7#15
FAQ
How much does MiMo VL 7B SFT 2508 cost?›
There is no per-unit API price: self-host only.
Is MiMo VL 7B SFT 2508 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is MiMo VL 7B SFT 2508, and what is it best at?›
#13 of 16 in Vision & Documents.
Is MiMo VL 7B SFT 2508 better than Qwen3 Omni 30B A3B Instruct?›
On Video-MME 2 · without subtitles/audio, MiMo VL 7B SFT 2508 scores 10.7 vs 10.9 — about the same.
What is the context window of MiMo VL 7B SFT 2508?›
128K tokens.
When was MiMo VL 7B SFT 2508 released? Is there a newer version?›
Released 2025-08-07. It is the current version in its line.
Is MiMo VL 7B SFT 2508 open source?›
Yes, the weights are open (MIT licence).
Can I run MiMo VL 7B SFT 2508 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can MiMo VL 7B SFT 2508 read images?›
It accepts text and images as input.
What is MiMo VL 7B SFT 2508?›
MiMo VL 7B SFT 2508 is an open-weights language model from Xiaomi, released Aug 2025.
What are the best alternatives to MiMo VL 7B SFT 2508?›
Gemini 3 Flash Preview (better), Qwen3 VL 8B Instruct (open weights).