Technology · LLMs → Fine-tuning · Open Source · wiki:stub
Axolotl
This pack is still a wiki stub. Opening text below if present; full article bands appear after deepen (`Wiki-status: deep`).
Axolotl is a popular open toolkit for fine-tuning LLMs with YAML configs covering LoRA, QLoRA, full FT, and many model families. Operationalizes domain specialization. Relevant when prompting alone cannot meet estimating/ops language needs. Go ahead and axolotl questions - 2026/08: - New model support has been added in Axolotl for Ling 3.0, Muse Glimmer, North Micro Vision Instruct and Shieldstral. - 2026/07: - NVFP4 (4-bit) MoE LoRA training is now supported via ScatterMoE (W4A16) and SonicMoE (W4A4), including adapter merge back into a plain NVFP4 checkpoint. - 2026/06: - Expert Parallelism (EP) for distributed MoE training via DeepEP, remote training through Tinker-compatible APIs, Context Parallelism for hybrid SSM models (Nemotron-H, Falcon-H1, Bamba), BitNet 1.58-bit fine-tuning, and a multimodal assistant-only loss-masking fix. - 2026/04: - New model support.
Research inventory
9 tags · 6 out · 7 in · 3 artifacts · 1 gaps · 14 corpus docs
Catalog tags
- landscape.layer
- LLMs
- landscape.subcategory
- Fine-tuning
- license_tag
- Open Source
- maps.dm
- absent
- maps.features
- 4
- one_liner
- LLMs
- review.depth
- science
- slug
- axolotl
- title
- Axolotl
Artifacts
- dm_map · absent
- features_map · present · technologies/axolotl/features.md
- readme · present · technologies/axolotl/README.md
Out · alternative_to
- llamafactorytechnology
- unslothtechnology
Out · maps_to
- bring-your-own-modelfeature
- learning-adaptationfeature
- model-flexibility-routingfeature
- r2-08-self-learningfeature
In · alternative_to
- llamafactorytechnology
- unslothtechnology
In · in_stack
- llms-fine-tuningstack · primary
In · maps_to
- bring-your-own-modelfeature
- learning-adaptationfeature
- model-flexibility-routingfeature
- r2-08-self-learningfeature
Corpus tags
- category
- LLMs → Fine-tuning
- dedication
- open-source
- feature
- bring-your-own-modellearning-adaptationmodel-flexibility-routingr2-08-self-learning
- kind
- map_edgetech_featurestech_quotetech_readme
- needs_deepen
- false
- quality
- ok
- slug
- axolotl
- source_id
- axolotl-githubllamafactory-contrastunsloth-contrast
- technology
- axolotl
Corpus documents (14)
map_edge · 4
- axolotl → bring-your-own-model
- axolotl → learning-adaptation
- axolotl → model-flexibility-routing
- axolotl → r2-08-self-learning
tech_features · 1
- Axolotl · features
tech_quote · 8
- Axolotl · axolotl-github
- Axolotl · axolotl-github
- Axolotl · axolotl-github
- Axolotl · llamafactory-contrast
- Axolotl · llamafactory-contrast
- Axolotl · llamafactory-contrast
- Axolotl · unsloth-contrast
- Axolotl · unsloth-contrast
tech_readme · 1
- Axolotl