—per task?
self-host onlyprice
—output speed?
66Kcontext
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-5.1unstated+10%—$1.25 / $10
OpenSkywork R1V3 38B-2%——Context: 66K → 33KBettero3unstated+7%—$2 / $8
BetterClaude Opus 4.5unstated+4%—$5 / $25dBetterDots.vlm1.instOpen+3%——Memory (4-bit): not verifiedModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-5.1unstated+10%—$1.25 / $10o3unstated+7%—$2 / $8Claude Opus 4.5unstated+4%—$5 / $25dDots.vlm1.instOpen+3%——Memory (4-bit): not verified
Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✗ 80 GB GPU
LicenceApache-2.0free for commercial use
Benchmarks
in Vision & DocumentsMMMU · validation77.7#9
FAQ
How much does Intern S1 cost?›
There is no per-unit API price: self-host only.
Is Intern S1 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Intern S1, and what is it best at?›
#9 of 41 in Vision & Documents.
Is Intern S1 better than Claude Sonnet 4.5?›
On MMMU · validation, Intern S1 scores 77.7 vs 77.8 — about the same.
What is the context window of Intern S1?›
66K tokens.
When was Intern S1 released? Is there a newer version?›
Released 2025-07-24. It is the current version in its line.
Is Intern S1 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Intern S1 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Intern S1 read images?›
It accepts text and images as input.
What is Intern S1?›
Intern S1 is an open-weights language model from internlm, released Jul 2025.
What are the best alternatives to Intern S1?›
GPT-5.1 (better), Skywork R1V3 38B (open weights).