Benchmark Hub

Ember-1

Fireworks · fireworks/ember-1

← Back to leaderboard

Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40%...

Ember-1 ranks #342 of 374 models in the default AI Benchmark Hub order (quality first, then cost). At $9.00 blended per million tokens, it is cheaper than 8% of models with published API pricing. Its 1,048,576-token context window is 4.0× the catalog median (262,144 tokens).

imagetexttext+image->text
vs Claude Opus 5vs Claude Fable 5vs Qwen3.8 27BTest in arenaOpen on OpenRouter

Context

Max context: 1048576
Max output: 943718

Pricing

Input / 1M: 3.00
Output / 1M: 15.00
Blend / 1M: 9.00

Quality

Quality index: —

Provider

Provider: Fireworks
Moderated: no

Monthly cost at this blended rate

Estimated API spend if every token is billed at $9.00per million (the blend of this model's input and output prices). Real bills depend on the input/output mix.

Tokens / monthEst. cost (USD)
1,000,000$9.00
10,000,000$90.00
100,000,000$900.00

Related models

Same lab, similar price, or similar quality — useful next comparisons.

Closest API price