{"version":1,"type":"story","url":"https://digestai.news/story/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model","json":"https://digestai.news/story/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model.json","markdown":"https://digestai.news/story/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model.md","slug":"fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model","headline":"Fireworks AI releases Ember-1, a post-trained Kimi K3 model","summary":"Fireworks AI has released Ember-1, a specialized model derived from Moonshot AI’s open-weight Kimi K3. The new model is post-trained to produce shorter reasoning traces while maintaining task accuracy, addressing the high token costs associated with reasoning models in multi-turn agentic workloads. Fireworks reports that Ember-1 delivers Kimi K3’s quality with approximately 40% fewer tokens. Unlike simply lowering the reasoning effort setting, which degrades quality, Ember-1 is trained to cut redundant reasoning loops while preserving useful self-reflection.\n\nThe model is currently available only as a Research Preview through the Fireworks serverless API. Fireworks has not released the weights, training code, or exact algorithms, meaning self-hosting is not possible. Pricing remains identical to Kimi K3 at $3.00 per million input tokens and $15.00 per million output tokens, with savings coming solely from reduced token generation.\n\nIn production A/B tests with two customers, Ember-1 reduced output tokens per task from 49.3K to 29.9K while keeping the task score nearly unchanged (0.753 vs 0.751). Fireworks states that Ember-1 leads Kimi K3 Max on Terminal Bench 2.1 and DeepSWE 1.1, though it trails slightly on SWE-bench Verified. The company claims these results were achieved using its own data and no customer data, with all training conducted on Fireworks Serverless Training.","keyPoints":["Ember-1 is a post-trained Kimi K3 model that uses about 40% fewer tokens while maintaining accuracy.","The model is available only via Fireworks serverless API as a Research Preview; weights are not released.","Production A/B tests showed output tokens per task fell from 49.3K to 29.9K with comparable quality scores."],"whyItMatters":"Ember-1 offers a cost-effective alternative for developers using reasoning models, significantly reducing inference costs for agentic workloads without sacrificing performance, though it remains locked behind a proprietary API.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["Fireworks AI","Moonshot AI","Doximity"],"models":["Ember-1","Kimi K3"],"people":["Asif Razzaq"]},"firstPublishedAt":"2026-09-28T07:22:33Z","updatedAt":"2026-09-28T07:22:33Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"MarkTechPost","title":"Fireworks AI Releases Ember-1: A Post-Trained Kimi K3 That Uses About 40% Fewer Tokens","url":"https://marktechpost.com/2026/09/28/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-that-uses-about-40-fewer-tokens","publishedAt":"2026-09-28T07:22:33Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Fireworks AI releases Ember-1, a post-trained Kimi K3 model\", 28 September 2026, https://digestai.news/story/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model","publisher":"Digest AI","title":"Fireworks AI releases Ember-1, a post-trained Kimi K3 model","datePublished":"2026-09-28T07:22:33Z","url":"https://digestai.news/story/fireworks-ai-releases-ember-1-a-post-trained-kimi-k3-model"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}