Feature/integrate llm providers - #28
Merged
Merged
Conversation
…, HuggingFace, etc.) Adds LLM_API_KEY/LLM_API_BASE_URL alongside the existing named providers (Anthropic, OpenAI, Gemini, Bedrock). When LLM_API_BASE_URL is set, requests route through CrewAI's native OpenAI-compatible client with the raw model id the endpoint expects; otherwise behavior is unchanged. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Add custom OpenAI-compatible endpoint alternative (NVIDIA NIM, Ollama, HuggingFace, etc.)
Contributor
Author
|
This should go to main branch to reflect all other llm providers |
3 tasks
aleksUIX
pushed a commit
to aleksUIX/seller-agent
that referenced
this pull request
Jul 26, 2026
… side) Policy: v2's architecture (service layer, per-domain routers) wins structurally; main's capabilities survive functionally. Resolutions: - src/ad_seller/interfaces/api/main.py: kept v2's app-assembly-only main. main's monolith additions (GAM reporting endpoints + rewired get_deal_performance, POST /products/search, placeholder POST /products/avails, hardened submit_proposal) are re-implemented in v2's service/router idiom in follow-up commits — except main's placeholder avails endpoint, which is superseded by v2's honest catalog-backed /products/avails and intentionally dropped. - src/ad_seller/interfaces/mcp_server.py: kept v2's thin-adapter registry; dropped main's httpx-loopback list_gam_orders / get_gam_delivery_report tools here (re-added as service-layer thin adapters in a follow-up commit, keeping the drift-guarded 44-tool inventory true at this commit). - src/ad_seller/tools/gam/book_deal.py: stays deleted (v2 EP-8.2 removed the abandoned CrewAI tools/ subpackages); its gam_order_id persistence intent moves to the service layer in the GAM re-wiring commit. - pyproject.toml: main's crewai[...,openai,gemini] extras don't exist on crewai>=1.15 — adopted origin/fix/crewai-google-genai-extra's spelling crewai[anthropic,bedrock,google-genai] (openai SDK is a core crewai dep). - Clean merges kept from main: src/ad_seller/llm/ (configurable providers, PR IABTechLab#28/IABTechLab#31) + tests/unit/test_llm.py, agents/crews build_llm adoption, GAMSoapClient reporting methods (PR IABTechLab#12), settings LLM+GAM fields, docs/guides/gam-reporting.md + configuration.md provider docs, ProposalHandlingFlow structured-output/async fixes (PR IABTechLab#29), serial product IDs in ProductSetupFlow. Suite after merge: 1128 passed, 26 skipped (baseline 1123 + 5 test_llm). bead: ar-rrkg Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KuoSai9uDgCEkbqYmxQpQ6
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.
Summary
docs/guides/configuration.md
Provider configuration (.env.example)
Only one provider can be active at a time. The file ships with Anthropic uncommented as the default. To switch, comment out the Anthropic block
and uncomment the target block:
--Provider: Anthropic (default — active)
ANTHROPIC_API_KEY=your-anthropic-api-key-here
DEFAULT_LLM_MODEL=anthropic/claude-sonnet-4-5-20250929
MANAGER_LLM_MODEL=anthropic/claude-opus-4-20250514
-- Provider: OpenAI (alternative)
--OPENAI_API_KEY=your-openai-api-key-here
--DEFAULT_LLM_MODEL=openai/gpt-4o
-- MANAGER_LLM_MODEL=openai/gpt-4o
--Provider: Google Gemini (alternative)
--GOOGLE_API_KEY=your-google-api-key-here
--DEFAULT_LLM_MODEL=gemini/gemini-2.5-flash
--MANAGER_LLM_MODEL=gemini/gemini-2.5-flash
Test plan