Omni-I2C (SVG) - Functional Coverage: leaderboard

Metric: Functional coverage S_func (0-100): share of the reference's pre-extracted, human-verified functional blocks present in the generated code, checked by a Claude Sonnet 4.5 parser against a fixed checklist, on the 192 SVG items; greedy decoding; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 13 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro76.3#77
2GPT-5.169.1#131
3Claude Sonnet 4.568.1#138
4Qwen 3 VL 235B A22B Instruct64.6#264
5Qwen 3 VL 32B Instruct64.2#276
6Qwen 3 VL 8B Instruct53.3#401
7Qwen 2.5 VL 72B Instruct50.4#364
8Gemma 3 27B (IT)44.1#509
9Qwen 2.5 VL 7B Instruct33.7#643

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=omni-i2c-svg-functional-coverage · How It Works · Data refreshed daily, snapshot 2026-10-11.