feat(providers): add Requesty as an OpenAI-compatible provider - #138
Closed
Thibaultjaigu wants to merge 1 commit into
Closed
feat(providers): add Requesty as an OpenAI-compatible provider#138Thibaultjaigu wants to merge 1 commit into
Thibaultjaigu wants to merge 1 commit into
Conversation
Requesty is an OpenAI-compatible LLM gateway using the same provider/model naming as OpenRouter, so this mirrors the existing OpenRouter provider on the generic openai_compat path. - core/providers/registry.py: requesty ProviderSpec (base URL https://router.requesty.ai/v1, REQUESTY_API_KEY, is_gateway) - core/providers/openai_compat.py: HTTP-Referer/X-Title attribution headers - new_ui/backend/services/requesty_models.py: models catalog service mapping Requesty's context_window / supports_* / flat price fields to the internal shape - new_ui/backend/api/routes/config.py: GET /config/requesty/models - core/config.py + deepcode_config.json.example + README provider list - tests mirroring the openrouter provider + models normalization Verified: ruff clean, 8 new tests pass; live completion through the provider code path and /v1/models both succeed against router.requesty.ai. Signed-off-by: Thibault Jaigu <thibault.jaigu@gmail.com>
Zongwei9888
added a commit
that referenced
this pull request
Aug 6, 2026
Ports three provider contributions that could not merge as submitted: #138 and #130 also touched new_ui/ and nanobot/, and #116 targeted nanobot/ only. Those trees were removed in the v2.0 refactor. The provider work itself applies unchanged to core/providers. ProvidersConfig reads each provider via getattr(providers, spec.name), so a ProviderSpec without a matching field is silently skipped — all three get both halves, as the ProvidersConfig docstring requires. Endpoints verified reachable: router.requesty.ai (200), api.minimax.io (401), api.forge.tensorblock.co (401). Also fixes a gateway bug #116 found in nanobot's provider that core had only half-fixed: spec.env_key was forced for gateways, but env_extras still used setdefault, so a stale ambient variable outranked the gateway the user picked. Dropped from the originals: the tests covering new_ui's requesty_models service and nanobot's registry. One of those imported "nanobot.config.schema", which now resolves to an unrelated PyPI package (nanobot-ai) when installed — it fails locally and skips in CI, testing nothing either way. Co-authored-by: Thibault Jaigu <thibault.jaigu@gmail.com> Co-authored-by: Yiiii0 <yiqiaozhou@outlook.com> Co-authored-by: octo-patch <octo-patch@github.com> Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Collaborator
|
Ported to Closing this branch rather than merging it because it also touched
Worth noting what sold this one: |
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
2 participants
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.
Adds Requesty as an OpenAI-compatible provider, mirroring the existing OpenRouter provider. Requesty is an LLM gateway that uses the same
provider/modelnaming convention as OpenRouter (e.g.openai/gpt-4o-mini,anthropic/claude-sonnet-4-5).Changes:
core/providers/registry.py— arequestyProviderSpecafter openrouter:openai_compatbackend,is_gateway, default base URLhttps://router.requesty.ai/v1, envREQUESTY_API_KEY.core/providers/openai_compat.py— forwardsHTTP-Referer/X-Titleattribution headers for Requesty (parallel to the OpenRouter path).new_ui/backend/services/requesty_models.py— models catalog service mirroringopenrouter_models.py, mapping Requesty'scontext_window,supports_tool_calling/supports_reasoningbooleans, and flatinput_price/output_priceto the internal model shape.new_ui/backend/api/routes/config.py—GET /config/requesty/models.core/config.py,deepcode_config.json.example,README.md— provider config + docs.tests/test_requesty_provider.py— mirrors the OpenRouter provider tests plus the models normalization.Testing:
ruff checkclean; 8 new tests pass and the existing model-compat tests still pass. Live-verified: a chat completion through the actual DeepCode provider path (make_llm_provider("requesty") -> OpenAICompatProvider) returned successfully againsthttps://router.requesty.ai/v1, and the models service normalized the live/v1/modelspayload (605 models).Note: the React model-selector UI (
new_ui/frontend) was left untouched to avoid shipping a partial frontend change; the backend/config/requesty/modelsroute is in place for a follow-up to wire the selector.Docs: https://requesty.ai · https://docs.requesty.ai · https://app.requesty.ai/api-keys · https://app.requesty.ai/router/list
I work at Requesty. This mirrors the existing OpenRouter provider as closely as possible. Happy to adjust or close it if it's not a fit.