Inception ships Mercury 2.5: diffusion LLM with 40% intelligence jump at 1,100 tokens/sec
timshi_ai · x · 2026-09-09
Inception Labs launched Mercury 2.5, claiming the most capable diffusion LLM on the market: a 40% intelligence jump over Mercury 2, running over 1,100 tokens/sec on widely available NVIDIA GPUs. It's live on the Inception API, OpenRouter, and Baseten.
Details: 260K context, reasoning, tool use, and structured output; intro pricing of $0.04/1M input and $0.15/1M output (80% off); claimed quality matching speed-optimized autoregressive models like GPT-5.2 and Claude Sonnet 4.5. Also announced: Mercury Voice for voice agents (TTFT under 170ms) and Mercury Router for prompt-based model routing.
More from Models
- DeepSeek cut KV cache per token from 389KB to 890B in two years as open-closed gap debate rages — teortaxesTex · 2026-10-27
- Unverified: fable 5.5 speculated at 4-5T parameters to outscale 1-3T Opus — haider1 · 2026-10-27
- Kimi K3 Is 'Open' Until You Succeed: License Caps Free Use at $20M Revenue — alex_verem · 2026-10-27
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Custom VQA benchmark under 48GB VRAM: Qwen3.8-Flash-Next leads, Qwen3.8-27B regressed — EmilPi · 2026-10-27