KAware - External Function Pass Rate: leaderboard
End-to-end task success on KAware (2026) tasks that can only be solved with external tools: the share of tasks whose trajectory and final answer an LLM judge marks correct, in percent; higher is better. KAware (2026) keeps tools available on every task, so an agent must judge whether it needs them.
Metric: The end-to-end pass rate (%): share of tasks whose trajectory and final answer a gpt-4.1 judge marks correct, on the external function setting, on the KAware tool-use tasks, where tools are always available but the task may need them (external function, 310 tasks), mix them with parametric knowledge (hybrid composition, 368 tasks) or be solvable from knowledge alone (internal function, 398 tasks); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B 2507 (Thinking) | 98.04 | #463 |
| 2 | Claude Sonnet 4.5 (Thinking) | 97.71 | #250 (Claude Sonnet 4.5) |
| 3 | Qwen 3 Max (Preview) | 97.1 | #447 |
| 4 | Qwen 3.5 397B A17B (Non-reasoning) | 96.13 | #246 (Qwen 3.5 397B A17B) |
| 5 | GPT-5.5 | 95.81 | #49 |
| 6 | Qwen 3.5 397B A17B | 93.87 | #246 |
| 7 | Claude Sonnet 4.5 | 93.55 | #250 |
| 8 | DeepSeek V4 Flash (Non-reasoning) | 92.26 | #258 (DeepSeek V4 Flash) |
| 9 | Gemini 3 Pro (Preview) | 88.06 | #113 |
| 10 | GPT-4.1 | 87.1 | #495 |
| 11 | Gemini 3 Flash (Preview) | 85.81 | #144 |
| 12 | GPT-4o (2024-11-20) | 85.16 | #866 |
| 13 | O4 Mini (2025-04-16) | 84.52 | #365 |
| 14 | GPT-5 | 84.19 | #167 |
| 15 | DeepSeek V3.2 (Non-reasoning) | 83.55 | #406 (DeepSeek V3.2) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=kaware-external-function-pass-rate · How It Works · Data refreshed daily, snapshot 2026-11-13.