• 0.5.0latestOpenAPI 3.0.3●18○32Docs2026-07-17 12:12
• Added adapter_source and api_token to completions API (#446)OpenAPI 3.0.3○4Docs16c3adc2024-04-29 22:59
• Make max_new_tokens optional, default to max_total_tokens - input_length (#353)OpenAPI 3.0.3○6Docs8ff0bf52024-03-22 17:01
• enh: Expose ignore_eos_token option in generate requests (#340)OpenAPI 3.0.3○3Docs67579cc2024-03-19 21:09
• Added Outlines logits processor for JSON schema validation (#224)OpenAPI 3.0.3○3Docs309618c2024-02-12 23:38
• Merge multiple LoRA adapters per request (linear, TIES, DARE) (#212)OpenAPI 3.0.3○3Docs3b4c9732024-02-01 05:39
• Fixed OpenAI stream response data (#193)OpenAPI 3.0.3●2○1Docs89062532024-01-19 06:11
• fix: OpenAI response format (#184)OpenAPI 3.0.3▲1●2○3Docs76fcc5d2024-01-16 16:48
• docs: OpenAI compatible API (#174)OpenAPI 3.0.3○2Docs64739ad2024-01-10 19:29
• OpenAI v1 Chat Completions API (#171)OpenAPI 3.0.3○3Docsa90d4432024-01-10 18:35
• Added `prompt_tokens` to the response (#165)OpenAPI 3.0.3○4Docsd88ffed2024-01-09 05:54
• Add predibase as a source for adapters (#125)OpenAPI 3.0.3○3Docs143f3032023-12-14 23:55
• Update openapi.jsonOpenAPI 3.0.3no API changesDocs41c657e2023-12-07 19:33
• Mkdocs (#112)OpenAPI 3.0.3▲9●18○41Docs284aaa22023-12-07 18:43

Revisions go back to the API's first capture.