7 commits

Author	SHA1	Message	Date
Yujong Lee	1a52bae779	add streaming message	2026-09-18 15:45:08 -07:00
Yujong Lee	c01db259d7	bring x-litellm-rust	2026-09-16 17:28:36 -07:00
Yujong Lee	c85acc8d28	test(rust_bridge): cover binding validation, async upstream errors, and OCR preparation failures ... Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>	2026-09-16 23:04:38 +00:00
Yujong Lee	62c862796a	cleanup	2026-09-16 14:38:59 -07:00
Yujong Lee	f9d423827f	fix(rust_bridge): let LITELLM_RUST win over litellm.rust() for optional tiers ... Parse the switch with pydantic TypeAdapter(bool) so 1/true/yes/on and 0/false/no/off all work, and treat an unparseable value as unset instead of off. PYTHON_ONLY and RUST_REQUIRED still ignore both switches Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>	2026-09-16 19:45:11 +00:00
Yujong Lee	7a1d433e7a	refactor(rust_bridge): declarative route catalog and shared runtime selection ... Replace the per-route enablement helpers (rust_enabled, rust_ocr_enabled, RUST_CHAT_COMPLETIONS_PROVIDERS, FallbackMode) with a single rule table in litellm/rust_bridge/catalog.py that maps a Context(route, provider, model, delivery) to one of four rollout tiers, and a pure decide() that turns tier plus process/env switches into a Decision. runtime.run/arun own the only fallback path: Python for PYTHON, native then Python on missing binding or admission decline for RUST_WITH_FALLBACK, raise for RUST_REQUIRED. OCR is the first route on the shared runtime; chat completions, Anthropic messages, and Responses websocket policy checks now read the catalog. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>	2026-09-16 19:10:42 +00:00
yujonglee	082bea851e	Merge pull request #39334 from BerriAI/litellm_rust_opt_in_configuration ... feat(python): unify Rust opt-in and bridge policy	2026-09-02 16:26:36 -07:00