3 Commits

Author	SHA1	Message	Date
huanghuoguoguo	9ecb587ac0	refactor(provider): use LiteLLM as unified LLM requester backend (#2150) ... * refactor(provider): use LiteLLM as unified LLM requester backend - Replace 23+ individual requester implementations with unified litellmchat.py - Add litellm_provider field to 27 YAML manifests for provider routing - Delete redundant requester subclasses - Add unit tests for LiteLLMRequester (29 tests) - Fix num_retries parameter name (was max_retries) - Fix exception handling order for subclass exceptions LiteLLM provides unified API for 100+ providers, eliminating need for provider-specific requesters. * fix: ruff format provider.py Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * refactor(provider): simplify LiteLLM requester usage handling - Remove unused Anthropic-specific tool schema generation - Share completion argument construction between normal and streaming calls - Use LiteLLM/OpenAI native usage fields for monitoring - Collect stream token usage from LiteLLM stream_options - Update LiteLLM requester tests for unified usage fields * restore: restore deleted provider requester files Restore individual provider requester implementations that were removed in de61b5d3. These files coexist with the unified litellmchat.py backend. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * feat: update requesters and improve provider selection UI - Added `litellm_provider` field to various requesters' YAML configurations. - Removed obsolete Python requester files for OpenRouter, PPIO, QHAIGC, ShengSuanYun, SiliconFlow, Space, TokenPony, VolcArk, and Xai. - Introduced new requesters for Tencent and Together AI with corresponding YAML configurations and SVG icons. - Enhanced the ProviderForm component to include a searchable dropdown for selecting providers, improving user experience. - Updated localization files to include search provider text for both English and Chinese. * fix(provider): align litellm rebase with master * fix(provider): capture streaming token usage; add token observability The LiteLLM streaming requester only captured usage when a chunk had an empty `choices` list. Many OpenAI-compatible gateways (e.g. new-api) and providers send the final usage payload in a chunk that still carries an empty-delta choice, so streamed calls always recorded 0 tokens in the monitoring logs/dashboard (non-streaming worked). - Capture stream usage whenever a chunk carries it, regardless of choices - Add robust _normalize_usage (dict/obj shapes, derive missing total_tokens) - Register litellm in bootutils/deps.py (was in pyproject only) - Add MonitoringService.get_token_statistics + /monitoring/token-statistics endpoint: summary, per-model breakdown, token timeseries, and a zero-token-success data-quality signal - Add TokenMonitoring dashboard tab (summary tiles, stacked token chart, per-model table) + i18n (en/zh) - Regression tests for stream usage capture and usage normalization Verified end-to-end against a real OpenAI-compatible endpoint with gpt-5.5 and claude-opus-4-8: tokens now recorded non-zero for both streaming and non-streaming paths. * refactor(provider): simplify litellm capabilities * style: simplify wrapped expressions * feat(models): persist context metadata * fix(provider): handle dict embeddings and openai-compatible rerank in LiteLLMRequester - invoke_embedding: support both object- and dict-shaped response.data entries (OpenAI-compatible gateways like new-api return dicts) - invoke_rerank: litellm.arerank rejects the 'openai' provider, so for openai-compatible (or unspecified) providers call the standard Jina/Cohere-style POST /v1/rerank endpoint directly over HTTP - accept both 'relevance_score' and 'score' fields in rerank results - add unit tests for the openai-compatible HTTP rerank path * feat(provider): enforce requester support_type when adding models - frontend: AddModelPopover only shows model-type tabs (llm/embedding/ rerank) that the provider's requester declares in its manifest support_type; ModelsDialog fetches requester manifests and maps requester -> support_type, passed down through ProviderCard - backend: add _validate_provider_supports guard in create_llm_model / create_embedding_model / create_rerank_model so a model cannot be attached to a provider whose requester does not support that type, even if the frontend restriction is bypassed (manifests without support_type are allowed for backward compatibility) - manifests: correct support_type for providers that do not offer all three model types: - llm only: anthropic, deepseek, groq, moonshot, openrouter, xai - llm + text-embedding: openai, gemini, mistral - add rerank to new-api (verified working via /v1/rerank) - set llm + text-embedding + rerank for aggregator/unknown gateways * feat(provider): add searchable alias to requester manifests - add a free-text 'alias' field to every requester manifest spec, containing the vendor's English/Chinese names, pinyin, common nicknames and flagship model-series names (e.g. moonshot -> kimi, 月之暗面; zhipu -> glm, 智谱清言) - frontend: ProviderForm requester search now also matches against alias (substring/contains), so searching 'kimi' surfaces Moonshot, '硅基' surfaces SiliconFlow, etc. - also fix support_type: openrouter (relay) supports embedding+rerank; LangBot Space gains rerank (coming soon) * fix(provider): make support_type guard defensive against incomplete model_mgr - _validate_provider_supports now uses getattr to gracefully skip when model_mgr / provider_dict / manifest lookup is unavailable, instead of raising AttributeError (fixes unit tests that mock ap.model_mgr as a bare SimpleNamespace) - add TestValidateProviderSupports covering: allow supported type, reject unsupported type, allow when support_type missing, allow when provider unknown, degrade safely when model_mgr is incomplete * fix(persistence): guard 0004 migration against missing llm_models table The 0004_add_llm_model_context_length migration called inspector.get_columns('llm_models') unconditionally, raising NoSuchTableError when the table does not exist (e.g. migrating a fresh/empty DB, as exercised by the integration tests where create_all() registers no tables because the ORM models are not imported). Every other migration guards with a table-existence check first; add the same guard here for both upgrade and downgrade. Also restore the test head assertion to 0004 (it had been lowered to 0003 to mask this failure). * Merge branch 'master' into feat/litellm Resolve conflicts: - uv.lock: regenerated via 'uv lock' to reconcile litellm/fastuuid (ours) with openai bump (master). - Alembic migrations: master added 0004_add_mcp_readme while this branch added 0004_add_llm_model_context_length, both as children of 0003 (would create multiple heads). Re-chain the litellm migration as 0005_add_llm_model_context_length with down_revision=0004_add_mcp_readme for a single linear head. Update test head assertion accordingly. * fix(persistence): shorten migration revision id to fit varchar(32) PostgreSQL stores alembic_version.version_num as varchar(32). '0005_add_llm_model_context_length' (33 chars) overflowed it, raising StringDataRightTruncationError in the PG migration tests. Rename the revision (and file) to '0005_add_llm_context_length' (27 chars) and update the head assertions in both SQLite and PostgreSQL migration tests. --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com> Co-authored-by: fdc310 <2213070223@qq.com> Co-authored-by: RockChinQ <rockchinq@gmail.com>	2026-06-13 16:59:48 +08:00
Junyan Chin	88132dff8a	perf: reduce memory usage by ~200MB+ at startup (#2013) ... * perf: reduce memory usage by ~200MB+ at startup Two key optimizations: 1. Use importlib.util.find_spec() instead of __import__() in dependency checking. find_spec() only locates modules without executing them, avoiding loading all 36 dependencies (~222MB) into memory at startup. 2. Introduce shared aiohttp.ClientSession via httpclient module. Previously, every HTTP request created a new ClientSession, which creates a new TCPConnector and SSL context, loading system root certificates each time (~270MB total allocations observed via memray). Now all HTTP client code reuses shared sessions. - satori.py and coze_server_api/client.py are left unchanged as they create one session per adapter lifecycle (not per-request). Profiling data (memray): - Peak memory: 403MB - SSL context creation: 270MB / 6.7M allocations (67% of total) - Dependency import: 222MB (55% of peak) - Expected reduction: 150-350MB at startup * fix: remove unused aiohttp imports (ruff F401) * style: ruff format	2026-02-27 20:09:03 +08:00
Copilot	e642ffa5b3	chore: Add PyPI package support for uvx/pip installation (#1764) ... * Initial plan * Add package structure and resource path utilities - Created langbot/ package with __init__.py and __main__.py entry point - Added paths utility to find frontend and resource files from package installation - Updated config loading to use resource paths - Updated frontend serving to use resource paths - Added MANIFEST.in for package data inclusion - Updated pyproject.toml with build system and entry points Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * Add PyPI publishing workflow and update license - Created GitHub Actions workflow to build frontend and publish to PyPI - Added license field to pyproject.toml to fix deprecation warning - Updated .gitignore to exclude build artifacts - Tested package building successfully Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * Add PyPI installation documentation - Created PYPI_INSTALLATION.md with detailed installation and usage instructions - Updated README.md to feature uvx/pip installation as recommended method - Updated README_EN.md with same changes for English documentation Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * Address code review feedback - Made package-data configuration more specific to langbot package only - Improved path detection with caching to avoid repeated file I/O - Removed sys.path searching which was incorrect for package data - Removed interactive input() call for non-interactive environment compatibility - Simplified error messages for version check Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * Fix code review issues - Use specific exception types instead of bare except - Fix misleading comments about directory levels - Remove redundant existence check before makedirs with exist_ok=True - Use context manager for file opening to ensure proper cleanup Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * Simplify package configuration and document behavioral differences - Removed redundant package-data configuration, relying on MANIFEST.in - Added documentation about behavioral differences between package and source installation - Clarified that include-package-data=true uses MANIFEST.in for data files Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> * chore: update pyproject.toml * chore: try pack templates in langbot/ * chore: update * chore: update * chore: update * chore: update * chore: update * chore: adjust dir structure * chore: fix imports * fix: read default-pipeline-config.json * fix: read default-pipeline-config.json * fix: tests * ci: publish pypi * chore: bump version 4.6.0-beta.1 for testing * chore: add templates/** * fix: send adapters and requesters icons * chore: bump version 4.6.0b2 for testing * chore: add platform field for docker-compose.yaml --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: RockChinQ <45992437+RockChinQ@users.noreply.github.com> Co-authored-by: Junyan Qin <rockchinq@gmail.com>	2025-11-16 19:53:01 +08:00