Ember-1 (Reasoning)
by Fireworks AI
Ember-1 (Reasoning) is a reasoning model from Fireworks AI, released on September 23, 2026. It has a 1M-token context window. API pricing is $3 per 1M input tokens and $15 per 1M output tokens. It ranks #36 of 220 models by Intelligence Score with a score of 43.1.
Ember-1 is a Fireworks Research specialized model built on Moonshot Kimi K3 that shortens reasoning traces (~40% fewer tokens) while matching K3-max quality on Fireworks' evaluations. Research Preview on Fireworks Serverless (model path accounts/fireworks/models/ember-1). MoE with 2.78T parameters, 1040k context, function calling. List pricing $3.00 input / $0.30 cached input / $15.00 output per 1M tokens (same as public Kimi K3 API pricing cited in the launch blog). Self-reported benches from the Fireworks blog: Terminal-Bench 2.1 82.0%, SWE-bench Verified 92.2%, SWE-Interact 20.0% (no catalog id — noted only), DeepSWE 1.1 75.2%, tau2-bench Airline 66%. Model page verified live (Ready) as of catalog add.
Key specs
| Provider | Fireworks AI |
|---|---|
| Release date | 2026-09-23 |
| License | proprietary |
| Context window | 1M tokens |
| Input price | $3 / 1M tokens |
| Output price | $15 / 1M tokens |
| Intelligence Score | 43.1 |
| Coding index | 37 |
Quick answers
What is the context window of Ember-1 (Reasoning)?
Ember-1 (Reasoning) has a context window of 1M tokens.
How much does Ember-1 (Reasoning) cost?
Ember-1 (Reasoning) costs $3 per 1M input tokens and $15 per 1M output tokens through the API.
Is Ember-1 (Reasoning) open source?
No. Ember-1 (Reasoning) is a proprietary model from Fireworks AI (proprietary license).
When was Ember-1 (Reasoning) released?
Fireworks AI released Ember-1 (Reasoning) on September 23, 2026.
How does Ember-1 (Reasoning) compare to Claude Opus 5.5 (Reasoning)?
Ember-1 (Reasoning) scores 43.1 on the Intelligence Score versus 60.5 for Claude Opus 5.5 (Reasoning), the current #1. Input pricing is $3 vs $4 per 1M tokens.