Ollama – v0.30.0-rc15

Ollama v0.30.0-rc15 🛠️

If you’re running LLMs locally, you know Ollama is the go-to for getting models like Llama 3 and DeepSeek-R1 up and running with zero friction. This latest release candidate brings a specific boost for Windows users looking to squeeze more performance out of their hardware!

What’s new:

• Windows iGPU Detection via Vulkan: The big highlight here is the improved detection of integrated graphics on Windows.

This is a massive win for anyone running Ollama on laptops or desktops without a dedicated high-end GPU. By better detecting available integrated graphics, Ollama can more effectively leverage your hardware’s compute power to speed up local inference. Keep an eye on those performance gains! 🚀

🔗 View Release

Ollama – v0.30.0-rc14: Merge remote-tracking branch ‘upstream/main’ into llama-runner-phase-0

Ollama v0.30.0-rc14 is officially in the works! 🛠️

If you haven’t jumped on the Ollama train yet, this is the go-to tool for running large language models (LLMs) locally on your machine with zero friction. It’s a total game-changer for devs who want to experiment with models like Llama 3 or Mistral without worrying about API costs or privacy leaks.

This specific release candidate focuses heavily on the internal plumbing and architectural refinement:

• Llama Runner Integration: This update marks a major milestone by merging the `upstream/main` branch into the `llama-runner-phase-0` development track.
• Core Optimization: The primary goal here is refining the runner architecture, which paves the way for more efficient model execution on your hardware.
• Workflow Stability: The release includes critical updates to the automated testing suites (`test.yaml`), ensuring that these heavy-duty runner changes don’t break your local setup.

Keep an eye on this one—as they bridge these branches, we can expect much smoother performance and better stability for local model execution! 🚀

🔗 View Release

Ollama – v0.30.0-rc13

Ollama v0.30.0-rc13 🛠️

If you’re running local LLMs, you know Ollama is the go-to for getting models like Llama 3 and DeepSeek-R1 up and running with zero friction. This latest release candidate is a focused update aimed at keeping your local inference engine sharp!

What’s new:

• llama.cpp Update: The big news here is an underlying update to the `llama.cpp` backend. Since Ollama relies on this for all the heavy lifting, these updates are huge for performance tweaks, improved memory management, and better support for the latest quantization methods. 🚀

Keep an eye on this one as it rolls out—backend refinements like this are exactly what we need to keep those local chats feeling snappy and efficient!

🔗 View Release

Ollama – v0.30.0-rc12

Ollama v0.30.0-rc12 🛠️

If you’re running local LLMs, you know Ollama is the gold standard for getting models like Llama 3 and DeepSeek-R1 up and running on your machine with zero friction. This latest release candidate (rc12) is a focused update aimed at polishing the experience!

What’s new in this release:

• Lint Fixes: The primary focus of this update is cleaning up the codebase. The developers have implemented several linting fixes to improve code quality, maintainability, and stability.
• Stability Improvements: By addressing these underlying issues, this release helps reduce potential bugs and ensures a smoother experience when pulling and running new models.

While it might not be a massive feature drop, these “under the hood” refinements are exactly what we love to see for keeping our local AI stacks reliable and production-ready! 🚀

🔗 View Release

Ollama – v0.30.0-rc11

Ollama v0.30.0-rc11 is here! 🛠️

If you’re running your own local LLM playground with models like Llama 3 or DeepSeek-R1, listen up! Ollama is the ultimate framework for getting powerful open-source models up and running on your machine without the cloud headache. This latest release candidate is all about polishing the experience and smoothing out those pesky edge cases.

What’s new in this update:

• Windows Optimization: Big win for my Windows tinkerer friends! 🪟 The team implemented a fix to prevent issues caused by spaces in the compiler path, making builds and installations much more stable on Windows environments.
• Stability Focus: Since this is an `rc` (release candidate) update, the primary goal is bug squashing. It’s designed to refine the stability of the current version before the full stable rollout hits your machine.

Keep testing these RC builds—they are the secret sauce to catching bugs before they hit your production workflows! 🚀

🔗 View Release

Ollama – v0.30.0-rc9

Ollama just dropped v0.30.0-rc9, and it looks like the team is focusing on polishing up the Windows experience! 🛠️

If you’re looking to run heavy hitters like Llama 3, DeepSeek-R1, or Mistral locally without relying on expensive cloud APIs, Ollama is your best friend. It handles all the heavy lifting of downloading and setting up models right on your machine.

What’s new in this release candidate:

• Windows Build Fix: The big news here is a fix for the Windows build assets. This update aims to ensure much smoother installations and more reliable builds for our fellow Windows power users. 🪟

It might seem like a small tweak, but stability is everything when you’re managing local inference! Keep an eye on these release candidates as they move toward the final stable version. 🚀

🔗 View Release

Ollama – v0.30.0-rc8: Merge remote-tracking branch ‘upstream/main’ into llama-runner-phase-0

Ollama v0.30.0-rc8 🦙

If you love running heavy-hitting models like Llama 3, DeepSeek-R1, or Mistral right on your own hardware, keep your eyes peeled! Ollama is making some serious moves under the hood to prep for the next generation of local LLM performance.

This release candidate is all about synchronization and stability as the team merges the latest upstream updates into the `llama-runner-phase-0` branch. Here’s the breakdown:

• Core Engine Sync: The development branch is being synced with the main upstream codebase. This is a huge step toward ensuring stability for upcoming runner features!
• Configuration Refinement: Updates to `envconfig/config.go` mean smoother handling of environment variables and runtime configurations. ⚙️
• Hardened Architectures: New updates to integration tests (specifically for model architectures and context windows) mean the engine is getting much more robust at handling complex, diverse models.

It’s a “behind-the-scenes” kind of update, but it’s exactly what we need to see for smoother, more reliable local workflows! 🛠️

🔗 View Release

Lemonade – v10.4.0: Fix ollama tool calling (#1780)

🍋 Lemonade SDK v10.4.0 is officially here!

If you’ve been looking for a way to squeeze every bit of performance out of your local hardware—especially leveraging NPUs and GPUs—Lemonade is the toolkit you need. It brings high-performance, private LLM serving right to your Windows or Linux machine with OpenAI API compatibility.

This latest update is a massive win for anyone building AI apps that rely on Ollama for local model execution. We’re seeing much better stability and much clearer communication between the SDK and your local engines.

What’s new in v10.4.0:

• Enhanced Model Visibility: You can now pull specific model capabilities and context window sizes directly via the Ollama API. No more “context window guesswork” when building your prompts! 🧠
• Fixed Tool Calling Logic: We’ve squashed a major bug where tool calls were being sent in streaming mode, which was causing malformed responses on the client side. Your agents should behave much better now.
• Protocol Alignment: Fixed a mismatch between the Ollama protocol and the OpenAI-compatible protocol used by `llama-server`. This ensures requests forwarded through Lemonade are no longer rejected as malformed.

Pro Tip: This patch was specifically tested with Zed v1.0.0. If you use Zed’s agent chat with built-in Ollama support, your coding workflow just got a whole lot more reliable! 🛠️

🔗 View Release

Ollama – v0.30.0-rc7

Ollama just dropped v0.30.0-rc7, and it looks like the team is fine-tuning things for an even smoother local LLM experience! 🛠️

If you’re looking to run heavyweights like Llama 3, DeepSeek-R1, or Mistral directly on your hardware without a massive cloud budget, this is the tool you need in your stack. This latest release candidate is all about stability and polishing the engine before the official stable rollout.

Here’s what’s new in this update:

• OpenMP Optimization: The team has implemented a change to disable OpenMP. This is a big deal if you’ve been running into threading conflicts or stability issues during model execution—it helps prevent clashes with other parallel processing libraries on your machine.
• Final Testing Phase: As an `rc7` build, this version is in the home stretch of bug-squashing. It’s the perfect time to test it out and see if these tweaks resolve any crashes you’ve been seeing during long inference sessions. 🚀

🔗 View Release

Ollama – v0.30.0-rc6

Ollama v0.30.0-rc6 🛠️

If you’re running local LLMs, you know Ollama is the go-to for getting models like Llama 3 and DeepSeek-R1 up and running with zero friction. This latest release candidate is a targeted update focused on stability for our Windows users!

What’s new:

• Windows Fix: The primary update in this release addresses an issue with dependency gathering on Windows systems. 🪟

This is a great little patch to keep your local inference engine running smoothly without those pesky missing dependency errors during setup or updates. Keep tinkering! 🚀

🔗 View Release