The open‑source Ollama model server now lets each reply include up to ten web lookups, and updates its MLX and llama.cpp libraries.
NoteOossa1 min read
Ollama released version 0.35.1‑rc0 on Sep 29, 2026. The build lets a model run up to ten web searches while generating a single answer. It also bumps the MLX library and the llama.cpp engine to newer commits. The change adds an option to declare a model’s explicit capabilities. These tweaks aim to improve answer accuracy and keep the software current.
More web lookups let Ollama produce fresher, fact‑based answers without leaving the local server.
Models
OpenAI’s new GPT‑6.1 Sol keeps the $2‑$10 per‑million‑token rate while claiming stronger performance, undercutting Anthropic’s Claude and DeepSeek models.
Models
The new v4 and v4 Turbo voices cost $0.08 and $0.04 per 1,000 characters, with a promotional cut to $0.022 and $0.011 per 1,000 until Oct 12.
Models
Oracle’s new Fusion Claw runtime lets its Fusion apps handle multi‑step jobs like staffing and accounting, using AI for reasoning while keeping calculations outside the model.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
| # | Source | Outlet | Date | Key takeaway |
|---|---|---|---|---|
| 1 | ollama/ollama v0.35.1 ↗ | Ollama | Sep 29, 2026 | ## What's Changed * feat: allow ten web searches per response by @ParthSareen in https://github.com/ollama/ollama/pull/18602 * MLX: version |
1 sources
