FrameBench - Frame Identification - English: leaderboard
Frame identification companion task from FrameBench (arXiv 2609.03370v1, Table 3), English: given a sentence, a target word and candidate frame names with definitions from FrameNet, the model selects the frame the word evokes. Accuracy in percent; higher is better. Evaluated by the FrameBench authors; reasoning and non-reasoning runs are separate rows.
Metric: Accuracy (%; FrameNet candidate frames). Source: arxiv.org. Saturation forecast: Around February 2027. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 84 | #105 |
| 2 | Gemini 3.1 Pro (Preview) | 82.5 | #64 |
| 3 | Gemma 4 31B (IT) (Thinking) | 82 | #178 (Gemma 4 31B (IT)) |
| 4 | Qwen 3.5 9B (Thinking) | 81.3 | #391 (Qwen 3.5 9B) |
| 5 | Qwen 3.5 4B (Thinking) | 81 | #514 (Qwen 3.5 4B) |
| 6 | Gemma 4 31B (IT) (Non-reasoning) | 81 | #178 (Gemma 4 31B (IT)) |
| 7 | Qwen 3.5 27B (Thinking) | 80.8 | #192 (Qwen 3.5 27B) |
| 8 | Gemma 4 E4B (IT) (Thinking) | 80.8 | |
| 9 | GPT-5 Nano | 77.8 | #439 |
| 10 | Qwen 3.5 27B (Non-reasoning) | 74.8 | #192 (Qwen 3.5 27B) |
| 11 | Gemma 4 E2B (IT) (Thinking) | 74.8 | #779 (Gemma 4 E2B (IT)) |
| 12 | Gemma 4 E2B (IT) (Non-reasoning) | 74 | #779 (Gemma 4 E2B (IT)) |
| 13 | Gemma 4 E4B (IT) (Non-reasoning) | 71.5 | |
| 14 | Qwen 3.5 2B (Thinking) | 68 | #1114 (Qwen 3.5 2B) |
| 15 | Qwen 3.5 9B (Non-reasoning) | 64 | #391 (Qwen 3.5 9B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=framebench-frame-identification-english · How It Works · Data refreshed daily, snapshot 2026-11-05.