0
0
Fork 0
mirror of https://github.com/discourse/discourse.git synced 2026-08-09 21:45:25 +08:00
discourse/plugins/discourse-ai/spec/lib/completions
Sam c9145ca98c
FIX: Improve vLLM completion handling (#41862)
Select the dedicated vLLM dialect before the OpenAI-compatible fallback.
Preserve response IDs on tool calls so parallel calls share provider
batch metadata, and account for vLLM's top-level prompt cache token
fields.

---------

Co-authored-by: discourse-patch-triage[bot] <272280883+discourse-patch-triage[bot]@users.noreply.github.com>
2026-07-21 16:35:05 +10:00
..
dialects FIX: Improve vLLM completion handling (#41862) 2026-07-21 16:35:05 +10:00
endpoints FIX: Improve vLLM completion handling (#41862) 2026-07-21 16:35:05 +10:00
anthropic_message_processor_spec.rb FEATURE: Provider-native built-in tools for agents (web search) (#40809) 2026-06-16 14:37:51 -03:00
cancel_manager_spec.rb FEATURE: retry transient and rate-limited LLM completion requests (#41237) 2026-06-30 09:45:26 +10:00
doc_to_text_spec.rb FEATURE: extract text from document uploads for LLM prompts (#39634) 2026-05-05 08:16:23 +10:00
docx_to_text_spec.rb FEATURE: extract text from document uploads for LLM prompts (#39634) 2026-05-05 08:16:23 +10:00
json_stream_decoder_spec.rb DEV: Enable Style/RedundantParentheses rubocop rule (#40095) 2026-05-19 15:48:09 +02:00
llm_metric_spec.rb
llm_presets_spec.rb FEATURE: Support OpenAI GPT-5.6 models (#41832) 2026-07-20 19:19:11 +10:00
llm_spec.rb FEATURE: retry transient and rate-limited LLM completion requests (#41237) 2026-06-30 09:45:26 +10:00
native_tools_spec.rb FEATURE: Provider-native built-in tools for agents (web search) (#40809) 2026-06-16 14:37:51 -03:00
ods_to_text_spec.rb FEATURE: extract text from ODT and ODS document uploads (#39711) 2026-05-05 12:04:13 -03:00
odt_to_text_spec.rb FEATURE: extract text from ODT and ODS document uploads (#39711) 2026-05-05 12:04:13 -03:00
open_ai_message_processor_spec.rb FEATURE: Support OpenAI GPT-5.6 models (#41832) 2026-07-20 19:19:11 +10:00
open_ai_responses_message_processor_spec.rb FEATURE: Support OpenAI GPT-5.6 models (#41832) 2026-07-20 19:19:11 +10:00
prompt_messages_builder_spec.rb DEV: Make AI context compaction global (#41317) 2026-07-14 18:41:37 -03:00
prompt_spec.rb FEATURE: Provider-native built-in tools for agents (web search) (#40809) 2026-06-16 14:37:51 -03:00
report_spec.rb FEATURE: add cost limits to LLM quotas (#41277) 2026-06-30 15:59:24 +10:00
rtf_to_text_spec.rb FEATURE: extract text from document uploads for LLM prompts (#39634) 2026-05-05 08:16:23 +10:00
structured_output_spec.rb FIX: Silent truncation and corruption of AI structured output streams (#41716) 2026-07-15 12:00:24 -03:00
token_usage_tracker_spec.rb DEV: Make AI context compaction global (#41317) 2026-07-14 18:41:37 -03:00
tool_arguments_parser_spec.rb FEATURE: Add AI authoring to Discourse Workflows (#40504) 2026-06-18 18:32:30 +02:00
tool_definition_spec.rb
upload_encoder_spec.rb FEATURE: extract text from ODT and ODS document uploads (#39711) 2026-05-05 12:04:13 -03:00
xls_to_text_spec.rb FEATURE: extract text from document uploads for LLM prompts (#39634) 2026-05-05 08:16:23 +10:00
xlsx_to_text_spec.rb FEATURE: extract text from document uploads for LLM prompts (#39634) 2026-05-05 08:16:23 +10:00
xml_tag_stripper_spec.rb
xml_tool_processor_spec.rb DEV: Enable Style/RedundantParentheses rubocop rule (#40095) 2026-05-19 15:48:09 +02:00