Fireworks Ember-1 Release - τ²-Bench Airline: leaderboard
Fireworks Research Ember-1 launch post (2026-09-23): controlled four-arm evaluation on the 50-task τ²-Bench Airline customer-service tool-use suite comparing Ember-1 (built on Kimi K3; reasoning effort unreported/default) against Fireworks' runs of Kimi K3 at reasoning efforts low, high and max (default). User simulator, scaffold and repeat count are not disclosed in the post. Pass Rate (%), higher is better.
Metric: Pass Rate (%). Source: fireworks.ai. Saturation forecast: Around February 2027. 2 models tracked. Updated 19 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Ember-1 | 66 | #92 | |
| 2 | Kimi K3 | Max | 64 | #61 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=fireworks-ember-1-release-tau2-bench-airline · How It Works · Data refreshed daily, snapshot 2026-11-20.