github unslothai/unsloth v0.1.3-beta
+50% tool call accuracy

5 hours ago

We did many new improvements and fixes to Studio!

  • Tool calls for all models are now +30% to +80% more accurate.
  • Web search now actually gets web content and not just summaries
  • Number of tool calls allowed are increased to 25 from 10
  • Tool calls now terminate much better, so looping / repetitions will be reduced
  • More tool call healing and de-duplication logic to stop tool callings from leaking XML as well
  • Tested with unsloth/Qwen3.5-4B-GGUF (UD-Q4_K_XL), web search + code execution + thinking enabled.
Metric Before After
XML leaks in response 10/10 0/10
URL fetches used 0 4/10 runs
Runs with correct song names 0/10 2/10
Avg tool calls 5.5 3.8
Avg response time 12.3s 9.8s

New features

  • Update button now visible
  • Install script styling all updated!
  • Added custom folders so you can use any GGUFs in any folder - for now access in Advanced Settings in Chat and Custom Folders
  • Preliminary Automatic Multi GPU support for inference and training - useful for large models that don't fit on 1 GPU - Studio auto will allocate GPU resources
  • Intel Macs should work out of the box

Much smoother and faster Studio

  • Fixed timeouts of downloads of large models - no more timeouts seen.
  • Fixed Hugging Face rate limiting - HF API calls reduced by 90%
  • Fixed bun on Windows and faster installs

To update Studio:

  1. For Linux, WSL, Mac, do: unsloth studio update
  2. For Windows native, do: irm https://unsloth.ai/install.ps1 | iex
  3. For Linux, WSL, Mac reinstalls, do: curl -fsSL https://unsloth.ai/install.sh | sh

What's Changed

Full Changelog: v0.1.25-beta...v0.1.3-beta

Don't miss a new unsloth release

NewReleases is sending notifications on new releases.