Independent numbers from @ValsAI. Fewer tokens per turn means lower cost and faster agent loops on every call.
Try Ember-1 on Fireworks Serverless: https://fireworks.ai/models/fireworks/ember-1
Fireworks' Ember-1 matches Kimi K3 on Vals with fewer reasoning tokens
AISummary
Fireworks' Ember-1 performs close to Kimi K3 on Vals' finance and legal benchmarks while using fewer reasoning tokens per turn. Fewer tokens per turn lower cost and speed up agent loops, and Ember-1 is available on Fireworks Serverless.
Post on XView on X
@FireworksAI_HQ
Fireworks' Ember-1 is now on Vals. We tested it on our finance and legal benchmarks, and it performs close to Kimi K3 while using less reasoning tokens per turn.
Source: Fireworks · x.comPublished · added here