Editorial review: Codex (AI-assisted source review), 2026-10-16.

Unsloth (September 2026 updates)

September updates to Unsloth's local model toolkit, covering skills, model support, Docker and multi-user operation.

Unsloth · Data, training, and fine-tuning platforms (datasets, labeling, RAG/vector DBs, fine-tuning-as-a-service) · tool

Published in Feed

Overview

Unsloth's September 2026 changelog documents Docker and multi-user updates on 17 September, followed by Qwen-Image-2.1 support and custom Agent Skills on 22 September. The latter update also covers project organisation, model handling and training exports. This entry covers those documented September changes rather than claiming to describe every current Unsloth capability.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Aymorphic analysis

Pin the toolkit and model versions together before comparing results. A shared workspace also needs a test that one account cannot retrieve another account's files or conversations.

Limitations

Changelog review only. Previous version-number and export-format claims were removed because this review did not establish them.

Key features

  • Docker and multi-user updates
  • Custom Agent Skills
  • Qwen-Image-2.1 support

Use cases

  • Maintaining a local model workspace
  • Reusing task instructions across sessions

Pricing

Confirm software licence, model licences and compute costs for the chosen deployment.

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
Not listed
Record updated
16 Oct 2026
Record last checked
16 Oct 2026

Explore the Feed

Other entries filed under data, training, and fine-tuning platforms (datasets, labeling, rag/vector dbs, fine-tuning-as-a-service); inclusion does not imply equivalent features.

Browse the Feed