SWE-bench Multimodal C
AI benchmark · Coding
What is SWE-bench Multimodal?
JavaScript issue resolution where bug reports depend on screenshots, evaluated by Vals AI with no agent internet access.
Tier C
Niche: a handful of models scored, or built by an RL vendor.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 17 models scored
- in curated RL directories (2 pts)
On X · poorly regarded, moderate buzz, from 6 posts about SWE-bench Multimodal in the last month
Vals launched a leaderboard, but the most-shared posts show models score the same or better without screenshots, so the benchmark doesn't test vision.
How we got to tier C
- Adoption: 17 models scored. Tier B.
- On X: poorly regarded, moderate buzz. “Vals launched a leaderboard, but the most-shared posts show models score the same or better without screenshots, so the benchmark doesn't test vision.” Moved down to C.
- In curated RL directories: 2 points, a floor of C.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Not on the map yet: no line from its own site or paper places it in a domain.
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-11-05 snapshot.