PulseAugur
EN
LIVE 14:48:43

Ollama releases v0.32.0 and v0.32.1 with new agent and Gemma 4 improvements · 2 sources tracked

Ollama has released two updates, v0.32.0 and v0.32.1, introducing significant new features and improvements. Version 0.32.0 launched an interactive agent experience designed to assist with coding and task delegation, and renamed the Codex App integration to ChatGPT. The subsequent update, v0.32.1, enhanced Gemma 4's tool-calling capabilities and multi-turn reasoning, while also addressing an MLX model cache leak and improving cache performance. AI

IMPACT Ollama's updates enhance user interaction and model capabilities, potentially improving developer productivity and the performance of integrated AI models.

RANK_REASON Software release for an AI tooling company.

Read on Mastodon — sigmoid.social →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Ollama releases v0.32.0 and v0.32.1 with new agent and Gemma 4 improvements · 2 sources tracked

COVERAGE [2]

  1. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    ⚙️ New Ollama Release! ⚙️ Version: v0.32.0 Release Notes: ## What's Changed - New interactive agent experience: running `ollama` now launches an agent to help y

    ⚙️ New Ollama Release! ⚙️ Version: v0.32.0 Release Notes: ## What's Changed - New interactive agent experience: running `ollama` now launches an agent to help you code and delegate work ``` ❯ ollama Ollama 0.32.0 ▸ Chat, Code, & Work (glm-5.2:cloud) Chat with models, code, search…

  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    ⚙️ New Ollama Release! ⚙️ Version: v0.32.1 Release Notes: ## What's Changed - Improved Gemma 4 tool calling and multi-turn reasoning, including more reliable to

    ⚙️ New Ollama Release! ⚙️ Version: v0.32.1 Release Notes: ## What's Changed - Improved Gemma 4 tool calling and multi-turn reasoning, including more reliable tool-response continuations - Fixed a recurrent MLX model cache leak that could increase memory use across requests, and i…