ERQA
Glimpse leads the ERQA leaderboard on BenchLM's November 2026 update with 83.8%, ahead of Qwen3.8 Max (77.8%) and Qwen3.8-Flash-Next (72.3%), across 19 models.
A grounded visual reasoning benchmark focused on evidence-based question answering over real images.
Top models on ERQA — November 21, 2026
As of November 21, 2026, Glimpse leads the ERQA leaderboard with 83.8% , followed by Qwen3.8 Max (77.8%) and Qwen3.8-Flash-Next (72.3%).
Glimpse
Elorian
Reported by ElorianQwen3.8 Max
Alibaba
Reported by AlibabaQwen3.8-Flash-Next
Alibaba
Reported by Alibaba19 modelsMultimodal & Grounded3% of Multimodal reference weightCurrentUpdated November 21, 2026
Leaderboard (19 models)
None of the 19 results on this page has a documented run by the benchmark or an independent evaluator: 12 reported by the labs that built the models and 7 reported in another publisher’s document.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | GlimpseElorianReported by Elorian | 83.8% | Reported by Elorian | Not reported | Closed |
| 2 | Qwen3.8 MaxAlibabaReported by Alibaba | 77.8% | Reported by Alibaba | Not reported | Open |
| 3 | Qwen3.8-Flash-NextAlibabaReported by Alibaba | 72.3% | Reported by Alibaba | Not reported | Open |
| 4 | Seed 2.1 ProByteDanceReported by ByteDance | 72.0% | Reported by ByteDance | Not reported | Closed |
| 5 | Seed 2.1 TurboByteDanceReported by ByteDance | 71.3% | Reported by ByteDance | Not reported | Closed |
| 6 | Qwen3.8-Omni-FlashAlibabaReported by Alibaba | 71.0% | Reported by Alibaba | Not reported | Closed |
| 7 | Gemini 3 ProGoogleAnother lab’s comparison table | 70.5% | Another lab’s comparison table | Not reported | Closed |
| 8 | Qwen3.7 PlusAlibabaReported by Alibaba | 69.8% | Reported by Alibaba | Not reported | Closed |
| 9 | Gemini 3.1 ProGoogleReported by Meta AI (comparison table) | 69.4% | Reported by Meta AI (comparison table) | Not reported | Closed |
| 10 | Qwen3.5 397BAlibabaReported by Alibaba | 67.5% | Reported by Alibaba | Not reported | Open |
| 11 | Qwen3.8-27BAlibabaReported by Alibaba | 65.5% | Reported by Alibaba | Not reported | Open |
| 12 | GPT-5.4OpenAIReported by Meta AI (comparison table) | 65.4% | Reported by Meta AI (comparison table) | Not reported | Closed |
| 13 | Muse SparkMetaReported by Meta | 64.7% | Reported by Meta | Not reported | Closed |
| 14 | Qwen3.6-27BAlibabaReported by Alibaba | 62.5% | Reported by Alibaba | Not reported | Open |
| 15 | GPT-5.2OpenAIAnother lab’s comparison table | 59.8% | Another lab’s comparison table | Not reported | Closed |
| 16 | Grok 4.20xAIReported by Meta AI (comparison table) | 54.1% | Reported by Meta AI (comparison table) | Not reported | Closed |
| 17 | Claude Opus 4.6AnthropicReported by Meta AI (comparison table) | 51.6% | Reported by Meta AI (comparison table) | Not reported | Closed |
| 18 | LFM3-VL-2BLiquidAIReported by LiquidAI | 48.2% | Reported by LiquidAI | Not reported | Open |
| 19 | Claude Opus 4.5AnthropicAnother lab’s comparison table | 46.8% | Another lab’s comparison table | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
According to BenchLM.ai, Glimpse leads the ERQA benchmark with a score of 83.8%, followed by Qwen3.8 Max (77.8%) and Qwen3.8-Flash-Next (72.3%). The scores show moderate spread, with meaningful differences between the top tier and mid-tier models.
19 models have been evaluated on ERQA. The benchmark falls in the Multimodal & Grounded category. BenchAlign v6.0 gives ERQA 3% of the Multimodal reference weight, so it moves the Multimodal leaderboard. Reference weights are relative weights in the calibrated model, not fixed shares of a score.
About ERQA
Year
2026
Tasks
Evidence-based visual QA
Format
Grounded image reasoning
Difficulty
Grounded multimodal reasoning
ERQA is useful as a grounded reasoning check because it emphasizes answer correctness tied to visual evidence rather than fluent but ungrounded descriptions.
Freshness and provenance
Version
ERQA 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 ERQA measure?
A grounded visual reasoning benchmark focused on evidence-based question answering over real images.
Which model scores highest on ERQA?
Glimpse by Elorian currently leads with a score of 83.8% on ERQA.
How many models are evaluated on ERQA?
19 AI models have published results on ERQA in the BenchLM catalog.
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 7,000+ subscribers.
One email each week. Unsubscribe anytime.