Flint by Recursive Now Free in Command Code With 262K Context
Command Code added Flint by Recursive as a free model on all plans, featuring a 262K token context window and a 40B-parameter architecture with 2.5B active para…
Command Code added Flint by Recursive as a free model on all plans, featuring a 262K token context window and a 40B-parameter architecture with 2.5B active para…
Lightpanda has built Jev into its browser binary to handle page triage, cookie walls, and logins in about 0.2 seconds, with chat models setting goals while Ligh…
Jev charges input tokens as state plus the sum of all questions, so state is billed only once regardless of how many parallel questions you ask. This makes it c…
OpenAI says ChatGPT for Healthcare now includes more capable computer-use agents, stronger clinical answers, and over 20 new healthcare plugins, with partners l…
GMI Cloud announces that Qwen4-Max will soon be available on its platform. The model uses a new MoE architecture with native vision and a 1M-token context windo…
Command Code now lets users assign a model and reasoning effort level to each subagent, so the main session can run on one model while subagents run in parallel…
OpenRouter data shows Fable 5.2 reached 11% of Anthropic model spend after one week, close to Fable 5.1 (12%) and Fable 5 (13%), while Opus 5.5 holds 46% wallet…
Hugging Face has integrated OpenRouter directly into its platform, allowing users to use the same token and endpoint, with live model stats on the Hub and a red…
OpenRouter and Hugging Face launched a partnership that routes Hugging Face Hub chat models through OpenRouter's provider selection, covering 160 models from 38…
In a study of more than 1,800 participants in the ChatGPT for Academic Researchers program, scientists reported nearly doubling their productivity, using privac…