MMSearch-Plus
We show this table for reference; we do not rank on it.
A harder MMSearch variant for multimodal retrieval and grounded tool-use workflows.
Benchmark score on MMSearch-Plus — September 27, 2026
We compile the MMSearch-Plus rows from provider self-reports. Qwen3.7 Plus leads the table at 41.4%. We do not use these results to rank models overall.
1 modelMultimodal & GroundedCurrentDisplay onlyUpdated September 27, 2026
Benchmark score table (1 model)
ScoreAbout MMSearch-Plus
Year
2026
Tasks
Hard multimodal search tasks
Format
Advanced mixed-media retrieval benchmark
Difficulty
Advanced multimodal search
BenchLM tracks MMSearch-Plus as a display-only extension of multimodal search capability.
Freshness and provenance
Version
MMSearch-Plus 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does MMSearch-Plus measure?
A harder MMSearch variant for multimodal retrieval and grounded tool-use workflows.
Which model scores highest on MMSearch-Plus?
Qwen3.7 Plus by Alibaba currently leads with a score of 41.4% on MMSearch-Plus.
How many models are evaluated on MMSearch-Plus?
1 AI models have been evaluated on MMSearch-Plus on BenchLM.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 2,000+ readers.
One email each week. Unsubscribe anytime.