Ember-1

Ember-1 is a specialized reasoning model from Fireworks Research, built on Kimi K3. It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40% fewer tokens than the base model while maintaining comparable quality across Fireworks' evaluations. It is suited for coding, knowledge work, and agentic workflows where reasoning cost and latency matter.

ChatFireworks1M tokens$3.2702 / $16.3509 · 1M

fireworks/ember-1

Context
1M tokens
Completion cap
943,718
Tools
Yes
JSON
Yes
Released
2026-09-24

Call it from Geek Hub

Same OpenAI SDK. Change the base URL and the model id.

curl https://api.geekhub.mx/v1/chat/completions \
  -H "Authorization: Bearer $GEEKHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "fireworks/ember-1",
    "messages": [{"role": "user", "content": "Hola"}]
  }'
Get an API key

Frequently asked questions

What is Ember-1?
Ember-1 is a specialized reasoning model from Fireworks Research, built on Kimi K3. It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40% fewer tokens than the base model while maintaining comparable quality across Fireworks' evaluations. It is suited for coding, knowledge work, and agentic workflows where reasoning cost and latency matter. Ember-1 runs on the Geek Hub API (OpenAI-compatible). Model id: fireworks/ember-1.
Is Ember-1 free?
No. Input is $3.2702 per 1M tokens and output is $16.3509 per 1M tokens on Geek Hub (markup included).
What is the context length of Ember-1?
Ember-1 has a 1M tokens context window. It supports up to 943,718 completion tokens.
Does Ember-1 support tool calling and structured outputs?
Ember-1 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
When was Ember-1 released?
Ember-1 was released on 2026-09-24.