Pokee-Isaac 28B Report - BFCL v4 - Live Simple: leaderboard
Berkeley Function-Calling Leaderboard v4 Live Simple category: a single call to a user-contributed real-world function. Score (%). From Pokee AI's Pokee-Isaac 28B technical report (September 10, 2026 revision): Pokee ran every listed model itself through hosted APIs under one harness, with each peer's default temperature (unless a benchmark section fixes one) and a high thinking level where available; exhausted retries, context overflows and verifier failures count as zero. Higher is better.
Metric: Score (%). Source: pokee.ai. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 122B A10B | 89.5 | #329 |
| 2 | Gemini 3.5 Flash Lite | 89.5 | #300 |
| 3 | Claude Haiku 4.5 | 86.8 | #568 |
| 4 | Nemotron 3 Super | 85.7 | #766 |
| 5 | GPT-5.6 Luna | 73.3 | #117 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=pokee-isaac-28b-report-bfcl-v4-live-simple · How It Works · Data refreshed daily, snapshot 2026-11-12.