Fireworks AI model product
Ember 1 is a Fireworks Research model built on Moonshot Kimi K3 that uses shorter reasoning traces while matching K3-max quality on Fireworks' evaluations.
Updated Oct 23, 2026. Default version: Ember-1
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Ember-1.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkSWE-Bench Verified | Score92.20% | Rank03 | Participants120 | Percentile98.32% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score66.00% | Rank08 | Participants26 | Percentile72.00% | EvidenceC | Evaluated |
| BenchmarkDeepSWE 1.1 | Score75.20% | Rank09 | Participants51 | Percentile84.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score82.00% | Rank22 | Participants48 | Percentile55.32% | EvidenceC | Evaluated |
Preference and agent-evaluation results for the default version.
The default version does not have a matching Arena result yet.
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKilo Gateway | Provider model IDfireworks/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDfireworks/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderFireworks AI | Provider model IDaccounts/fireworks/models/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDfireworks/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDfireworks/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCloudflare AI Gateway | Provider model IDfireworks/ember-1 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Ember-1. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionEmber-1 | Released | LLMBoard71.07 | Parameters2.8T | Context1M | Max outputN/A | Open weightsNo | LicenseProprietary |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Ember 1 and its available data.
78T parameters, a 1040k context window, and function calling. It is available as a Research Preview on Fireworks Serverless under accounts/fireworks/models/ember-1.
Data as of 2026-10-22.
Common questions about Ember 1.
Ember 1's default version was released on Sep 23, 2026.
Ember 1's official API price is $3 per million input tokens and $15 per million output tokens via Fireworks AI. The lowest tracked third-party offer starts at $3 input and $15 output via Kilo Gateway.
Ember 1 was created by Fireworks AI.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
6 provider offerings are linked to the default version.