Ember-1 (distilled from Kimi K3) tested: ~20% cheaper, 36% less reasoning tokens, 2x faster
sophiamyang · x · 2026-10-02
User kickingkeys noted that Ember-1, derived from Kimi K3, still shows traces of its base model in its paintings but expresses ideas in its own way. sophiamyang shared her testing across 124 prompts: Ember-1 costs 20% less, spends 36% less on reasoning tokens, and is 2x faster than the base model.
Related event: Ember-1 Tested Faster and Cheaper Than Kimi K3(2 posts)→
More from Models
- Opus 5.5 keeps saying "himbo" — a verbal quirk no previous Claude showed — repligate · 2026-10-02
- AI spend falls in latest Ramp AI Index as frontier price cuts bite; open source under 5% — PaulYacoubian · 2026-10-02
- Mystery model Fledge Alpha spotted before announcement, clues point to Thinking Machines Lab — cephaloform · 2026-10-02
- Ivo's contract agent hits 91% on Legal Agent Benchmark via River AI post-training — ibab · 2026-10-02
- Gemini 4 Argon called 'frontier' yet ranks #48 in analysis and #59 in presentation — AbjectMycologist614 · 2026-10-02
- Artificial Analysis adds refusal timing and fallback-model views to Coding Agent Index — ArtificialAnlys · 2026-10-02