multimodal benchmark
MMLongBench-128K
MMLongBench-128K evaluates multimodal long-context understanding at a 128K token context length, testing how well vision-language models reason over very long mixed text and image inputs.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
MMLongBench-128K leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 78.3% | 100.0% | 2 | C | |
| 2 | BY | 76.9% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
MMLongBench-128K highlights
The top published results on this benchmark's own scale.
What is MMLongBench-128K?
Definition and scoring fields from the benchmark registry.
MMLongBench-128K evaluates multimodal long-context understanding at a 128K token context length, testing how well vision-language models reason over very long mixed text and image inputs.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- MMLongBench-128K
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mmlongbench-128k|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about MMLongBench-128K.
Which model scores highest on MMLongBench-128K?
Seed 2.1 Pro is currently ranked first with 78.3%.
What does MMLongBench-128K measure?
MMLongBench-128K evaluates multimodal long-context understanding at a 128K token context length, testing how well vision-language models reason over very long mixed text and image inputs.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.