Fireworks AI has announced significant cost reductions for its GLM-5.2 model, claiming it is approximately 48% cheaper than Anthropic's Opus 4.7. The company achieved this by reducing the cached token price for GLM-5.2 by 2X and launching a post-training version with zero-KLD. Fireworks AI states that GLM-5.2 costs $1.12 per session, compared to Opus 4.7's $2.14 per session, when normalized to a 90% cache hit rate. AI
IMPACT This announcement may pressure competitors to lower pricing for similar inference services.
RANK_REASON This is a cost comparison and product update from a specific AI infrastructure provider, not a frontier model release or major industry event.
Read on X — Fireworks (inference infra) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
