ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
unsloth/studio/backend/assets/configs
History
Exact
Octopus 9beefcaf7c
 feat(studio): expose provider_type selector in model provider dialog (#4277) 
* feat(studio): expose provider_type selector in model provider dialog

Use the configured provider_type in the model provider payload instead of hardcoding openai. Add a provider type selector to the model provider dialog for the provider types supported by Data Designer.

* feat(studio): add MiniMax-M2.7 inference defaults

Add MiniMax-M2.7 model family entry to inference_defaults.json
with recommended parameters (temperature=1.0, top_p=0.95, top_k=40).
Pattern placed before minimax-m2.5 for correct longest-match-first
ordering.

* Fix/adjust provider type support for PR #4277

* Clarify provider type validation for PR #4277

* Fix provider type default validation for PR #4277

---------

Co-authored-by: wasimysaid <wasimysdev@gmail.com>
Co-authored-by: PR Bot <pr-bot@minimaxi.com>
2026-06-15 15:07:06 +02:00
..
model_defaults Add Gemma 4 model sampling defaults (#4838) 2026-04-03 13:57:15 -07:00
__init__.py Final cleanup 2026-03-12 18:28:04 +00:00
full_finetune.yaml fix(studio): correct default weight_decay and learning rate (#4695) 2026-03-31 13:50:25 +04:00
inference_defaults.json feat(studio): expose provider_type selector in model provider dialog (#4277) 2026-06-15 15:07:06 +02:00
lora_text.yaml fix(studio): correct default weight_decay and learning rate (#4695) 2026-03-31 13:50:25 +04:00
vision_lora.yaml fix(studio): correct default weight_decay and learning rate (#4695) 2026-03-31 13:50:25 +04:00