11 Commits

Author SHA1 Message Date
The BROKE Cluster Team bf7480d042 Release 2.0.4-beta.9: Audio transcription via mlx-audio
Major Features:
- Audio transcription via mlx-audio backend (Whisper, >10min duration)
- OpenAI /v1/audio/transcriptions endpoint
- Memory Gate System (Vision: 8GB, Audio: 4GB)
- Config-based backend routing (ADR-020)
- Benchmark toolchain (memmon/memplot, Schema v0.2.2)

Key Fixes:
- EuroLLM tokenizer decoding
- Vision-model text-only routing regression
- Multimodal model context length detection
- Memory cleanup bug (mx.metal.clear_cache)
- Orphan process bug

Test Results:
- Unit tests: 647 passed, 11 skipped (Python 3.10-3.12)
- wet-umbrella: 171 passed total

See CHANGELOG.md for complete details and known issues.
2026-02-04 03:10:30 +01:00
The BROKE Cluster Team 5751545b8b Release 2.0.4-beta.7: Server robustness + Vision per-chunk streaming
- Server: exit codes, /v1/models crash fix, vision routing, MLXK2_MAX_TOKENS
- Vision: true SSE streaming, hallucination fix (local numbering)
- Workspace: list prefix-match, push ambiguous pattern handling
- Docs: SERVER-HANDBOOK accuracy updates

See CHANGELOG.md for details.
2026-01-18 16:57:32 +01:00
The BROKE Cluster Team 53d9cca82d Release 2.0.4-beta.6: Local workspace workflow + Vision batch processing
- Complete local development cycle: clone → repair → run/show/server on
  workspace paths without HuggingFace round-trips
- Vision processing now defaults to safe chunking (one image at a time,
  prevents OOM + hallucination)
- Resumable clone with --force-resume and deterministic temp cache naming
- Improved test infrastructure (umbrella marker convention)
- 161 Wet Umbrella tests passing including new Vision→Geo pipe integration tests

See CHANGELOG.md for complete details.
2026-01-07 17:11:07 +01:00
The BROKE Cluster Team 25609e4dcb Release 2.0.4-beta.5: Community repair tool + OS-agnostic benchmarking
Closes #49 (Mistral Tokenizer Bug)

Major features:
- Workspace Infrastructure (ADR-018 Phase 0a): Managed workspace detection,
  provenance metadata, backward compatible with unmanaged workspaces
- Convert Operation (ADR-018 Phase 1): `mlxk convert --repair-index` fixes
  mlx-vlm #624 affected models (7+ models including Qwen2.5-VL, gemma-3)
- Resumable Pull: Auto-detect partial downloads with `--force-resume`
- Wet Umbrella Test Integration: Single entry point for all real model tests

Fixes:
- #49: BPE space markers now correctly converted (Mistral-family models)
- Vision Portfolio Discovery: Filter by capabilities instead of model_type
- Memory Cleanup Hook: Triggers for both live_e2e and wet markers

Test suite: 528 passed, 60 skipped (Python 3.9-3.14)
2025-12-31 16:05:18 +01:00
The BROKE Cluster Team 86f669dc82 Release 2.0.4-beta.1: Vision + Pipes + Memory
- Vision Support (Issue #45): CLI + Server with OpenAI-compatible image API, EXIF metadata
- Unix Pipes (ADR-014): stdin support, isatty detection, SIGPIPE handling
- Memory-Aware Loading (ADR-016): Pre-load checks with >70% RAM warnings
- Python 3.9-3.14: Full compatibility verified (476-485 tests passing)
- Fixed: --log-json regression (Issue #44), Vision multimodal history filtering

See CHANGELOG.md for complete details.
2025-12-16 19:35:30 +01:00
The BROKE Cluster Team 57bf6d86be 2.0.0-beta.3: Feature Complete - Full 1.1.1 Parity Achieved
Major Features Added:
  • Complete run command implementation with interactive/single-shot modes
  • MLXRunner core engine ported from 1.x with modular architecture
  • OpenAI-compatible server with SIGINT-robust supervisor mode
  • Experimental push feature properly isolated behind environment variable

  Key Improvements:
  - Full feature parity with 1.1.1 stable releases
  - Enhanced human output formatting across all commands
  - Clean separation of stable (184 tests) vs experimental features
  - Updated demo GIF showcasing improved 2.0 interface

  Fixes:
  - Pull operation cache pollution (Issue #30) with preflight access checks
  - Test stability improvements across all environments

  Architecture:
  - Modular runner design with focused helper modules
  - Thread-safe model loading and memory management
  - stable testing across Python 3.9-3.13

  Ready for use as comprehensive 1.x alternative.
2025-09-14 18:04:18 +02:00
Local Test 19a66674c0 2.0.0-alpha.1: human output default; strict health (#27, PyTorch index)
See CHANGELOG.md and README.md
2025-08-31 22:25:43 +02:00
Local Test 2624210353 Consolidate JSON API; add --version; health in list/show (spec json-0.1.2)
- Enforce strict JSON API in tests
- Introduce --version --json as sole version reporter
2025-08-31 15:17:19 +02:00
Local Test de7ccf9018 2.0.0-alpha: default 2.0 tests, cache safety, and docs
Testing:
- pytest defaults to tests_2.0 via pytest.ini
- README/TESTING updated; Quick Start uses `pip install -e . && pip install pytest`

Safety:
- Add test-cache sentinel + centralized checks
- Strict delete guard via MLXK2_STRICT_TEST_DELETE=1
- Hide sentinel from 2.0 list output

Portability:
- Remove site-specific paths; generic test/user cache detection (mlxk2_test_ prefix + sentinel)

Docs:
- Environment & Caches, HF cache integrity
- Local-only hooks/excludes and local test script (excluded from VCS)
2025-08-29 16:57:45 +02:00
The BROKE Team d375e1bd3e MLX-Knife 2.0.0-alpha: Issue #27 Discovery & Development README
Major Achievements:
- Live reproduction and documentation of Issue #27 (health check false positive)
- Comprehensive development README.md for alpha phase parallel usage
- JSON API specification integration and references
- 45/45 tests passing with production-quality reliability

Issue #27 Critical Discovery:
- Health check false positives for multi-part model downloads
- Root cause: Multi-part pattern detection flaw in shared logic
- GitHub issue created with reproduction steps and technical analysis

2.0.0-Alpha Development Status:
- Revolutionary test isolation architecture complete
- Atomic cache system with triple safety verification
- Development handbook with parallel deployment guide
- Ready for production testing and broke-cluster integration
2025-08-28 23:49:14 +02:00
The BROKE Team c5777a3e7a MLX-Knife 2.0 Session 2 Complete + Session 3 Major Progress: Production-Ready Alpha
Session 2 Achievements (100% Complete):
• All 5 operations implemented: list, health, show, pull, rm
• CLI compatibility: fuzzy matching, @hash syntax, short names
• Lock detection: New safety feature replacing interactive prompts
• Pull corruption recovery: Clean rm→pull workflow
• Hash fuzzy matching: show "model@3df9bfd" works with short hashes

Session 3 Major Progress (70% Complete):
• JSON API Specification v0.1.1: Complete and production-ready
• Full SHA hashes: 40-char hashes in list output for broke-cluster
• Show command: --files and --config options fully implemented
• JSON-flag enforcement: --json required in alpha for correct usage

Key Features:
• Broke-cluster integration ready with enforced --json flag
• Real-world tested: rm→pull workflow validated with corrupted cache
• Lock cleanup: 9 lock files properly cleaned during rm operations
• Sanitized JSON: No cache paths or implementation details exposed

Testing Status: Basic unit tests implemented, manual validation successful
Note: Test suite requires update for JSON API Specification v0.1.1

Package: Installable as mlxk-json alongside 1.1.0

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-27 21:36:35 +02:00