Tooling11d ago
LightOn open-sources ColGrep-Agent-2B: local 2B code-search agent cuts Claude Code API tokens ~34% and cache-aware cost ~40%
On 2026-10-12 LightOn AI released ColGrep-Agent-2B (fine-tuned from MiniCPM5-2B, Apache-2.0): a local 2B code-localization agent that searches with ColGrep multi-vector retrieval plus a read-only terminal, then returns file paths and line ranges to a frontier coding agent. On 52 code-search questions across 13 unseen repos (624 Claude Code / Sonnet 5.5 sessions), delegating search cut API tokens ~34.3% and cache-aware cost ~40% vs grep, with 80.8% vs 83.3% accuracy (not statistically significant). On SWE-bench Lite localization it posted 19.92 line F1—best among models evaluated in the same ColGrep harness, including several >10× larger. Ships brew/cargo/GGUF installers and hooks for Claude Code, Codex, and OpenCode.
Tooling39d ago
Gemini API lists antigravity-preview-09-2026: Antigravity agent endpoint, not a new model
The Gemini API now lists id antigravity-preview-09-2026. It is not a new Pro/Flash model but the Antigravity managed-agent endpoint: Linux sandbox with code execution, files, and browser, defaulting to Gemini 3.8 Flash. Google-affiliated Philipp Schmid confirmed it is the Gemini API agent harness, with more details expected later this week.
Release8d ago
Gemini 4 Argon is generally available: gemini-4-argon ships in the Gemini API and AI Studio with a 1M unified window, long decode and $2/$10 intro pricing
On Oct 13 Google made Gemini 4 Argon globally available: the stable gemini-4-argon model is live on the Gemini API paid tier and AI Studio, in Antigravity for AI Ultra, and in the Gemini app and AI Mode for Pro and Ultra subscribers. Input, thinking and output share a 1,048,576-token window, and background execution lets one response run up to 1M tokens. Intro pricing is $2 input / $10 output per million tokens through Nov 30, then $4 / $20. There is no free tier, explicit caching is unsupported, and temperature/top_p/top_k are ignored.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.