feat: add Requesty as an OpenAI-compatible provider - #449
Open
Thibaultjaigu wants to merge 1 commit into
Open
Thibaultjaigu wants to merge 1 commit into
Thibaultjaigu wants to merge 1 commit into
Conversation
Adds a Requesty provider family following the MiniMax/ZAI pattern (OpenAILike transport, base https://router.requesty.ai/v1) plus a load_llm branch next to OpenRouter and DeepSeek. Handles REQUESTY_API_KEY and REQUESTY_BASE_URL (for the EU, US and AP regional routers), seeds the model menu from Requesty managed policy ids, and documents the provider, env vars and regions.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds Requesty (one OpenAI-compatible API across 700+ models from OpenAI, Anthropic, Google, xAI and others) as a provider. It follows the existing MiniMax / ZAI pattern: a first-class provider family that reuses the
OpenAILiketransport, plus aload_llmbranch alongside OpenRouter and DeepSeek. That waymobilerun configureoffers Requesty with a base URL prompt and a model menu without any wizard changes, since the wizard is registry driven.Changes
mobilerun/agent/providers/requesty.py: base URLs (global, EU, US, AP),REQUESTY_API_KEYandREQUESTY_BASE_URLhandling viaresolve_requesty_base_url, default modelopenai/gpt-4o-mini, and a model list seeded from Requesty managed policy ids (GET /v1/models/managed, verified live on 2026-09-11). Full<vendor>/<model>catalog ids work as well.registry.py:requestyProviderFamilySpecand env key slot.env_keys.py:REQUESTY_API_KEY.llm_picker.py:RequestyinSUPPORTED_PROVIDERS,requestyalias, dispatch branch that resolves the key, default model and base URL and routes toOpenAILike.cli/main.py: help strings for--provider,--base_urlandconfigure --provider.README.md,SKILL.md,docs/quickstart.mdx,docs/guides/cli.mdx(with a region table including the EU endpoint) anddocs/sdk/configuration.mdx.tests/test_requesty_configuration.py(17 tests mirroringtest_minimax_configuration.py) covering the registry entry, setup profile generation, key source resolution, picker defaults and base URL overrides.No new dependency: Requesty rides on the always installed
llama-index-llms-openai-like. Not added to any automatic provider preference.Validation
ruff checkandblack --checkclean on all touched files.test_legacy_minimax_profile_warns_once_without_migration, fails identically on cleanmainand is unrelated.load_llm("Requesty", model="openai/gpt-4o-mini")(themobilerun run --provider Requestypath) and theconfigureprofile path (apply_selection_to_rolestoload_llms_from_profiles) both completed a real chat call. Every model id in the menu was confirmed present in the live/v1/models/managedand/v1/modelscatalogs.Disclosure: I work at Requesty. Happy to adjust anything to match project conventions.