Fireworks Launches FireRouter: 98.1% of Opus Accuracy at 57% Lower Coding Cost
nicolechirps · x · 2026-09-29
Fireworks AI introduced FireRouter with Opus, its first cache- and task-aware routing model, available as a standalone serverless endpoint. It routes each user turn between Claude Opus 5.5, GLM 5.3, and GLM 5.3 Flash, weighing quality against cost including prompt-cache loss.
In a month-long internal A/B test on coding traffic:
- Cost per session fell 57% (±19 ppts), from $15.36 to $6.63
- Graded-turn accuracy hit 78.7% vs 80.2% for Opus-only, or 98.1% of Opus accuracy
- Scoring covered task completion, answer correctness, and whether users had to correct the next turn
More from coding & agent
- Hackathon project Beethoven turns paintings into a live AI band playing via Lyria — schwentker · 2026-09-29
- OpenAI DevDay agenda leaks: Codex to get platform capabilities for plugins, agents and apps — testingcatalog · 2026-09-29
- Concept car fully generated in code, zero assets, one HTML file, built with Claude Sonnet 5.5 — techartist_ · 2026-09-29
- LLMs Keep Stalling on Lead Enrichment: Lazy Output and Hallucinated Emails — Royal_icey69 · 2026-09-29
- Auditing Thousands of Rollouts: 80%+ of Coding Agents Reason About an Imagined Grader — jonas__m · 2026-09-29
- 120 FPS in Browser: A Universal Decompiler Steps Closer to Reality — yacineMTB · 2026-09-29