Fixing the ruff linting

Add OSS evaluation suite, universal JSON crush, latency benchmarks

Merge pull request #33 from angpt/angpt/any-llm-integration

Update CHANGELOG.md

Bump version to 0.3.7

Fix CI: guard starlette imports, asyncio.run(), deprecate datetime.utcnow()

Fix: guard starlette imports in test_compress_api.py for CI without proxy deps

README: replace plain text architecture with Mermaid diagrams

README: add back proof points — needle-in-haystack demo, benchmarks, architecture detail

README: proxy as the hero quickstart, compress() for Python, integration table for existing proxies

Rewrite README + add Integration Guide

Add one-function compress() API, ASGI middleware, LiteLLM callback

Add Compression Hooks — extension points for SaaS and advanced customization

Update anyllm.py

Bump version to 0.3.6

Add Query Echo: re-inject user question after compressed tool outputs

Add compression summaries, multi-provider headers, Dockerfile fix

Fixing the DockerFile to have hnswlib working

Add pluggable adapter hooks for CCR, Storage, and TOIN backends

Add tests for streaming resilience and concurrent session safety

Fix streaming crashes under concurrent sessions

Update anyllm.py

Bump version to 0.3.2

Add adaptive compression sizing and fix cost tracking bugs

Update anyllm.py

Adding any-llm as a backed provider

Fix security vulnerabilities in memory and CCR systems

Add LiteLLM backend routing for OpenAI endpoint and Magika content detection

Fix token savings metrics mismatch between pipeline and server

Fix inflated token savings stats from cache hits

Use AWS API to dynamically fetch Bedrock inference profiles

Fix MCP tests to work when MCP SDK is not installed

Add MCP CLI for Claude Code subscription users

Merge pull request #26 from prakersh/dev/prakersh/fix

fix: Handle BaseModelOutputWithPooling from SigLIP get_*_features()

Merge pull request #25 from prakersh/fix/increase-request-body-size-limit

test: Update LLMLingua default model test to match new default

fix: Increase MAX_REQUEST_BODY_SIZE from 10MB to 100MB to support image-heavy requests

Add centralized ML model configuration

Bump version to 0.3.1

Fix: Include dashboard HTML templates in wheel package

Fix mypy error: add type annotation for SentenceTransformer return

Fix test_router_is_available_with_models after MLModelRegistry refactor

Add centralized MLModelRegistry to share ML model instances

Fix test to expect VectorBackend.AUTO as default

Add SQLite vector backend as default and fix HNSW test skipping

Add SQLiteVectorIndex using sqlite-vec for bounded vector search

Add bounded memory support to HNSWVectorIndex with LRU eviction

Merge pull request #24 from chopratejas/feature/memory-observability

Add SQLiteGraphStore for bounded, persistent graph storage