InternVL3 5 30B A3B Instruct 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?
41Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGemini 3 Flash Previewunstated×2.6—$0.50 / $3OpenGemma 4 31Bunstated+51%—$0.10 / $0.34Memory (4-bit): 18 GB → 19 GB
BetterMiMo v2 Omniunstated×2.4—$0.14 / $0.28
BetterKimi K2.5unstatedOpen×2.2—$0.60 / $3Memory (4-bit): 18 GB → 616 GB
BetterGPT-5unstated×2.1—$1.25 / $10ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Gemini 3 Flash Previewunstated×2.6—$0.50 / $3MiMo v2 Omniunstated×2.4—$0.14 / $0.28Kimi K2.5unstatedOpen×2.2—$0.60 / $3Memory (4-bit): 18 GB → 616 GBGPT-5unstated×2.1—$1.25 / $10Gemma 4 31BunstatedOpen+51%—$0.10 / $0.34Memory (4-bit): 18 GB → 19 GBModelScoreCostPer taskWhat changes
Gemma 4 31BunstatedOpen+51%—$0.10 / $0.34Memory (4-bit): 18 GB → 19 GB
Qwen2.5 VL 72B InstructunstatedOpen+9%—$2.80 / $8.40Memory (4-bit): 18 GB → 44 GBQwen3 VL 8B InstructunstatedOpensame—$0.18 / $0.69MiMo VL 7B RL 2508Opensame——Qwen3 VL 4B InstructunstatedOpen-8%—$0.10 / $0.60ModelScoreCostPer taskWhat changes
OInternVL3 5 241B A28B InstructOpen+25%——Memory (4-bit): 18 GB → 141 GBOInternVL3 5 8B InstructOpen-10%——OInternVL3 5 4B InstructOpen-21%——OInternVL3 1BOpenn/a——Context: 41K → 33KOInternVL3 2BOpenn/a——Context: 41K → 33KOInternVL3 8BOpenn/a——Context: 41K → 33KAll OpenGVLab models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
in Vision & DocumentsVideo-MME 2 · with subtitles/audio16.6#14
Video-MME 2 · without subtitles/audio12.6#9
FAQ
How much does InternVL3 5 30B A3B Instruct cost?›
There is no per-unit API price: self-host only.
Is InternVL3 5 30B A3B Instruct free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is InternVL3 5 30B A3B Instruct, and what is it best at?›
#8 of 16 in Vision & Documents.
Is InternVL3 5 30B A3B Instruct better than Qwen2.5 VL 72B Instruct?›
On Video-MME 2 · without subtitles/audio, InternVL3 5 30B A3B Instruct scores 12.6 vs 13.7 — lower.
What is the context window of InternVL3 5 30B A3B Instruct?›
41K tokens.
When was InternVL3 5 30B A3B Instruct released? Is there a newer version?›
Released 2025-08-25. It is the current version in its line.
Is InternVL3 5 30B A3B Instruct open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run InternVL3 5 30B A3B Instruct locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can InternVL3 5 30B A3B Instruct read images?›
It accepts text and images as input.
What is InternVL3 5 30B A3B Instruct?›
InternVL3 5 30B A3B Instruct is an open-weights language model from OpenGVLab, released Aug 2025.
What are the best alternatives to InternVL3 5 30B A3B Instruct?›
Gemini 3 Flash Preview (better), Gemma 4 31B (open weights).