# ERQA leaderboard

Multimodal & Grounded benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/multimodal-grounded-erqa

Every model the catalog carries a published ERQA value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Multimodal & Grounded |
| Measure | Grounded image reasoning |
| Tasks | Evidence-based visual QA |
| Difficulty | Grounded multimodal reasoning |
| Published by | [Qwen3.6 launch benchmarks](https://qwen.ai/blog?id=qwen3.6) |

A grounded visual reasoning benchmark focused on evidence-based question answering over real images.

## ERQA ranking

19 models with a published ERQA value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Grounded image reasoning | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Glimpse](https://modelchoice.dev/models/glimpse) | Elorian | 83.8 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Qwen3.8 Max](https://modelchoice.dev/models/qwen3-8-max) | Alibaba | 77.8 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Qwen3.8-Flash-Next](https://modelchoice.dev/models/qwen3-8-flash-next) | Alibaba | 72.3 | Observed 2026-11-21 · source benchlm:models |
| 4 | [Seed 2.1 Pro](https://modelchoice.dev/models/seed-2-1-pro) | ByteDance | 72.0 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Seed 2.1 Turbo](https://modelchoice.dev/models/seed-2-1-turbo) | ByteDance | 71.3 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Qwen3.8-Omni-Flash](https://modelchoice.dev/models/qwen3-8-omni-flash) | Alibaba | 71.0 | Observed 2026-11-21 · source benchlm:models |
| 7 | [Gemini 3 Pro](https://modelchoice.dev/models/gemini-3-pro) | Google | 70.5 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Qwen3.7 Plus](https://modelchoice.dev/models/qwen3-7-plus) | Alibaba | 69.8 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Gemini 3.1 Pro](https://modelchoice.dev/models/gemini-3-1-pro) | Google | 69.4 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Qwen3.5 397B](https://modelchoice.dev/models/qwen3-5-397b) | Alibaba | 67.5 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Qwen3.8-27B](https://modelchoice.dev/models/qwen3-8-27b) | Alibaba | 65.5 | Observed 2026-11-21 · source benchlm:models |
| 12 | [GPT-5.4](https://modelchoice.dev/models/gpt-5-4) | OpenAI | 65.4 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Muse Spark](https://modelchoice.dev/models/muse-spark) | Meta | 64.7 | Observed 2026-11-21 · source benchlm:models |
| 14 | [Qwen3.6-27B](https://modelchoice.dev/models/qwen3-6-27b) | Alibaba | 62.5 | Observed 2026-11-21 · source benchlm:models |
| 15 | [GPT-5.2](https://modelchoice.dev/models/gpt-5-2) | OpenAI | 59.8 | Observed 2026-11-21 · source benchlm:models |
| 16 | [Grok 4.20](https://modelchoice.dev/models/grok-4-20-beta) | xAI | 54.1 | Observed 2026-11-21 · source benchlm:models |
| 17 | [Claude Opus 4.6](https://modelchoice.dev/models/claude-opus-4-6) | Anthropic | 51.6 | Observed 2026-11-21 · source benchlm:models |
| 18 | [LFM3-VL-2B](https://modelchoice.dev/models/lfm3-vl-2b) | LiquidAI | 48.2 | Observed 2026-11-21 · source benchlm:models |
| 19 | [Claude Opus 4.5](https://modelchoice.dev/models/claude-opus-4-5) | Anthropic | 46.8 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
