Luis Pater 690f4f3116 feat(executor): support use-max-completion-tokens for openai compatibility models ...

- Add `use-max-completion-tokens` setting to OpenAI compatibility model configuration
- Provide token normalization between `max_tokens` and `max_completion_tokens` based on model preference

Closes: #5939

2026-09-19 01:54:05 +08:00

..

helps

feat(executor): support use-max-completion-tokens for openai compatibility models

2026-09-19 01:54:05 +08:00

aistudio_executor_test.go

…

aistudio_executor.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

antigravity_executor_auth.go

…

antigravity_executor_buildrequest_test.go

feat(translator): enhance web search streaming and citation mapping in OpenAI Responses

2026-09-17 03:51:48 +08:00

antigravity_executor_compaction_test.go

…

antigravity_executor_credits_test.go

…

antigravity_executor_credits.go

…

antigravity_executor_disable_cooling_test.go

…

antigravity_executor_execute.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

antigravity_executor_finish_reason_test.go

…

antigravity_executor_interactions_test.go

…

antigravity_executor_keepalive_test.go

…

antigravity_executor_request.go

…

antigravity_executor_signature_test.go

…

antigravity_executor_split_usage_test.go

…

antigravity_executor_stream.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

antigravity_executor_tokens.go

feat(translator): preserve model metadata in requests

2026-09-17 13:10:47 +08:00

antigravity_executor_transport_test.go

…

antigravity_executor.go

feat(translator): enhance web search streaming and citation mapping in OpenAI Responses

2026-09-17 03:51:48 +08:00

antigravity_home_model_capabilities_test.go

…

antigravity_preupstream_rewrite_differential_test.go

…

antigravity_preupstream_rewrite_legacy_oracle_test.go

…

antigravity_reasoning_replay_clear_test.go

…

antigravity_reasoning_replay_index_test.go

…

antigravity_reasoning_replay_legacy_oracle_test.go

…

antigravity_reasoning_replay_test.go

…

antigravity_reasoning_replay.go

…

antigravity_refresh_test.go

…

antigravity_schema_sanitize_test.go

…

caching_verify_test.go

…

claude_cloaked_cache_repro_test.go

…

claude_executor_auth_race_test.go

…

claude_executor_auth_test.go

…

claude_executor_auth.go

…

claude_executor_beta_passthrough_test.go

…

claude_executor_beta_policy_test.go

…

claude_executor_cloaking.go

fix(executor): restrict claude advisor tool check to server tool use

2026-09-18 23:32:00 +08:00

claude_executor_diagnostics_test.go

…

claude_executor_diagnostics.go

…

claude_executor_execute.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

claude_executor_fable_ratelimit_test.go

fix(claude): skip retry-after header on overage-only rejections

2026-09-18 09:15:48 +08:00

claude_executor_fast_error_test.go

…

claude_executor_fast_error.go

…

claude_executor_native_helper_test.go

…

claude_executor_ratelimit_test.go

…

claude_executor_request_bench_test.go

…

claude_executor_request_remap_test.go

…

claude_executor_request.go

feat(claude): support model-level cooling and scope overage rate limits

2026-09-18 01:55:36 +08:00

claude_executor_stream_terminal_test.go

fix(executor): prevent stream failure on client disconnect after claude completion

2026-09-16 20:00:27 +08:00

claude_executor_stream.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

claude_executor_subagent_ttl_regression_test.go

…

claude_executor_test.go

fix(executor): restrict claude advisor tool check to server tool use

2026-09-18 23:32:00 +08:00

claude_executor_thinking_signature_test.go

…

claude_executor_tokens.go

feat(claude): support model-level cooling and scope overage rate limits

2026-09-18 01:55:36 +08:00

claude_executor_wire_casing_test.go

…

claude_executor.go

feat(claude): support model-level cooling and scope overage rate limits

2026-09-18 01:55:36 +08:00

claude_fingerprint_policy_test.go

…

claude_fingerprint_policy.go

…

claude_mid_system_model_test.go

…

claude_signing_test.go

…

claude_signing.go

…

claude_thinking_replay_test.go

…

claude_thinking_replay.go

…

codex_executor_auth.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

codex_executor_cache_test.go

…

codex_executor_compact_test.go

…

codex_executor_execute.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

codex_executor_grokbuild_keepalive_test.go

…

codex_executor_imagegen_test.go

…

codex_executor_input_ids_test.go

…

codex_executor_instructions_test.go

…

codex_executor_parallel_tool_calls_test.go

…

codex_executor_reasoning_replay_cache_test.go

…

codex_executor_reasoning.go

…

codex_executor_request.go

…

codex_executor_retry_test.go

…

codex_executor_signature_test.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

codex_executor_spawn_agent_test.go

…

codex_executor_stream_output_test.go

…

codex_executor_stream.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

codex_executor_terminal.go

feat(codex): record upstream response model and warn on silent model substitution

2026-09-18 12:31:52 +08:00

codex_executor_tokens.go

…

codex_executor_tool_schema_test.go

…

codex_executor_translate_test.go

…

codex_executor.go

…

codex_native_fidelity_test.go

…

codex_openai_images_extract_test.go

…

codex_openai_images_test.go

…

codex_openai_images.go

feat(codex): record upstream response model and warn on silent model substitution

2026-09-18 12:31:52 +08:00

codex_response_model_test.go

feat(codex): record upstream response model and warn on silent model substitution

2026-09-18 12:31:52 +08:00

codex_stream_bootstrap_buffering_test.go

…

codex_websockets_connection.go

…

codex_websockets_errors.go

…

codex_websockets_execute.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

codex_websockets_executor_store_test.go

…

codex_websockets_executor_test.go

…

codex_websockets_executor.go

…

codex_websockets_request.go

…

codex_websockets_session.go

…

codex_websockets_spawn_agent_test.go

…

codex_websockets_stream.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

custom_magic_headers_test.go

…

devin_executor_test.go

fix(executor): record expected Devin upstream model and bound stream observer memory

2026-09-18 22:08:35 +08:00

devin_executor.go

fix(executor): record expected Devin upstream model and bound stream observer memory

2026-09-18 22:08:35 +08:00

executor_payload_optimization_test.go

…

gemini_executor_signature_test.go

…

gemini_executor_test.go

fix(interactions): strip invalid id from function_result and redundant call_id from function_call

2026-09-16 00:31:31 +08:00

gemini_executor.go

fix(executor): record authentic Devin and Gemini Interactions response models

2026-09-18 21:32:59 +08:00

gemini_vertex_executor.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

home_codex_terminal_test.go

…

kimi_executor_test.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

kimi_executor.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

kimi_home_model_capabilities_test.go

…

kimi_thinking_replay_test.go

…

kimi_thinking_replay.go

…

meta_executor_execute.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

meta_executor_stream.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

meta_executor_test.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

meta_executor.go

…

openai_compat_executor_compact_test.go

…

openai_compat_executor_images_test.go

fix(executor): robust response model observability across meta, kimi, and openai-compat streams

2026-09-18 21:53:19 +08:00

openai_compat_executor_max_tokens_test.go

feat(executor): support use-max-completion-tokens for openai compatibility models

2026-09-19 01:54:05 +08:00

openai_compat_executor_reasoning_test.go

…

openai_compat_executor_retry_test.go

…

openai_compat_executor_tool_results_test.go

fix(executor): strip relayed tool result images for text-only models

2026-09-17 09:57:13 +08:00

openai_compat_executor.go

feat(executor): support use-max-completion-tokens for openai compatibility models

2026-09-19 01:54:05 +08:00

openai_responses_signature_test.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

openai_responses_signature.go

fix(codex): preserve reasoning content and IDs for compat models in responses

2026-09-18 23:02:54 +08:00

response_model_multiprovider_test.go

fix(executor): record authentic Devin and Gemini Interactions response models

2026-09-18 21:32:59 +08:00

websocket_lifecycle_bind_test.go

…

websocket_session_target_test.go

…

xai_executor_auth.go

…

xai_executor_execute.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

xai_executor_media.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

xai_executor_request.go

fix(xai): restore aliased client web search tool name in responses

2026-09-18 09:57:06 +08:00

xai_executor_response.go

fix(xai): restore aliased client web search tool name in responses

2026-09-18 09:57:06 +08:00

xai_executor_stream.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00

xai_executor_test.go

fix(xai): restore aliased client web search tool name in responses

2026-09-18 09:57:06 +08:00

xai_executor_tokens.go

…

xai_executor.go

fix(xai): restore aliased client web search tool name in responses

2026-09-18 09:57:06 +08:00

xai_reasoning_replay.go

…

xai_status_err_test.go

…

xai_websockets_executor_test.go

fix(xai): restore aliased client web search tool name in responses

2026-09-18 09:57:06 +08:00

xai_websockets_executor.go

feat(executor): extend response model recording and substitution warnings to all providers

2026-09-18 21:22:43 +08:00