Ollama 0.40.7-rc0 adds PPLX Decider support in MLX
Updated Ollama
Summary
A prerelease adds PPLX Decider v1.2 support and adjusts Responses compatibility and speculative-draft handling.
- Tool
- Ollama
- Category
- Agent infra
- Date
- Open source
- Yes
- License
- MIT
- Pricing
- Self-hosting, infrastructure and model-provider charges are separate; no hosted free allowance is promised.
Ollama published v0.40.7-rc0 on October 18, 2026. GitHub marks it as a prerelease; the release title alone should not be read as a stable-channel announcement.
What changed
The notes list three changes: PPLX Decider v1.2 support in the MLX backend, accepting Responses effort on non-thinking models, and allowing llama-server to infer the draft speculative type.
The MLX addition is the main new capability here. The other changes concern compatibility and draft configuration, not evidence that existing models became more accurate or faster. This release also does not establish the licensing or hardware requirements of every model a user might load.
Who should look at it
Users evaluating the newly supported model or affected API behavior can inspect the candidate before adopting it. Anyone needing a stable environment should distinguish this release candidate from the stable version already deployed in their own setup.
License and cost
Ollama's code license is MIT. That license does not cover every model's weights, and running locally still requires suitable hardware. The pricing page separately describes cloud plans and usage credits. Local execution is not an unlimited free cloud API.
See the Ollama hub.
More from Ollama
Ollama 0.40.6 adds Clef cloud support and q2_0 types
OllamaOllama’s October 16 release adds Clef in its cloud service, GGUF q2_0 types and a Windows runner cleanup fix.
Category: Agent infra
Pricing: Open source
Ollama 0.40.2 upgrades downloaded models for llama.cpp
OllamaOllama v0.40.2, released October 8, 2026, upgrades previously downloaded models in the background for llama.cpp and keeps backups for safe downgrades.
Category: Coding agents
Pricing: Open source
Questions
What is Ollama?
According to its site, Ollama gives access to the latest open models with privacy, locally or in the cloud. The repository describes it as a way to get up and running with models such as Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen and Gemma.
Is Ollama open source?
The GitHub API reports the ollama/ollama repository license as MIT.
What is the latest launch tracked here?
Ollama v0.40.2, published October 8, 2026, which upgrades previously downloaded models in the background for llama.cpp.
Filed on Agent Radar. Claims in this note are limited to the sources above.