kk573 c80b2e2c34 fix: use MaxCompletionTokens instead of deprecated MaxTokens for OpenAI (#1604) ...

The OpenAI API deprecated 'max_tokens' parameter in favor of
'max_completion_tokens' for newer models (o1, gpt-4o, etc.).

This change fixes the error:
'Unsupported parameter: max_tokens is not supported with this model.
Use max_completion_tokens instead.'

Refs: https://platform.openai.com/docs/api-reference/chat/create#chat-create-max_tokens

Signed-off-by: Evgenii Kuzakov <evgeniy.kuzakov@csssr.com>
Co-authored-by: Evgenii Kuzakov <evgeniy.kuzakov@csssr.com>

2026-01-27 17:46:51 +00:00

..

bedrock_support

chore: model name (#1535)

2025-06-20 16:30:50 +01:00

interactive

feat: interactive mode (#854)

2024-01-14 20:22:00 +00:00

amazonbedrock_mock_test.go

feat: add support for Amazon Bedrock Inference Profiles (#1492)

2025-05-06 11:18:40 +01:00

amazonbedrock_test.go

chore: model access (#1529)

2025-06-20 13:27:49 +01:00

amazonbedrock.go

feat: fix to broken inference (#1575)

2025-09-03 20:08:44 +01:00

amazonsagemaker.go

feat: implement Top-K sampling for improved user control (#1110)

2024-05-16 15:41:07 +01:00

azureopenai.go

feat: dump (#1322)

2024-11-12 12:04:01 +00:00

bedrock_interfaces.go

feat: add support for Amazon Bedrock Inference Profiles (#1492)

2025-05-06 11:18:40 +01:00

cohere.go

chore(deps): update cohere client implementation to v2 (#1062)

2024-04-20 21:10:17 +01:00

customrest.go

feat: add custom restful backend for complex scenarios (e.g, rag) (#1228)

2025-03-17 12:21:38 +00:00

factory.go

fix: config ai provider in query (#1457)

2025-04-15 11:11:40 +01:00

googlegenai.go

chore: fix typo in "completion" (#1362)

2025-02-20 11:48:52 +00:00

googlevertexai.go

feat: add latest and legacy stable models (#1539)

2025-06-27 08:34:23 +01:00

groq.go

feat: add Groq as LLM provider (#1600)

2026-01-05 12:05:17 +00:00

huggingface.go

feat: implement Top-K sampling for improved user control (#1110)

2024-05-16 15:41:07 +01:00

iai.go

feat: add Groq as LLM provider (#1600)

2026-01-05 12:05:17 +00:00

localai.go

feat: unify aiClientName const for all providers (#848)

2024-01-07 13:01:15 +01:00

noopai.go

feat: unify aiClientName const for all providers (#848)

2024-01-07 13:01:15 +01:00

ocigenai.go

feat: oci genai chat models (#1337)

2025-07-20 10:02:47 +01:00

ollama.go

feat: add Ollama backend (#1065)

2024-07-03 21:16:06 +01:00

openai_header_transport_test.go

feat: add custom http headers to openai related api backends (#1174)

2024-07-10 09:59:08 +01:00

openai.go

fix: use MaxCompletionTokens instead of deprecated MaxTokens for OpenAI (#1604)

2026-01-27 17:46:51 +00:00

prompts.go

feat: add custom restful backend for complex scenarios (e.g, rag) (#1228)

2025-03-17 12:21:38 +00:00

watsonxai.go

fix: rename watsonxai to ibmwatsonxai (#1234)

2024-09-24 21:58:01 +01:00