v0.8.2

What's Changed

• Better json detection and extraction from strings by @tjake in #105

Full Changelog: v0.8.1...v0.8.2

v0.8.1

What's Changed

• Add missing F32 Q4 operations for arm by @tjake in #103

Full Changelog: v0.8.0...v0.8.1

v0.8.0

What's Changed

• Adds Downloader class to add fluent coding when different options by @lordofthejars in #87
• chore: update JlamaRingWorkerService.java by @eltociear in #90
• Gemma2 classifier sample by @tjake in #92
• Add utf-8 charset to ui by @LuccaPrado in #96
• Fix qwen tokenizer since they use more than one by @tjake in #99
• Add support for granite models by @mariofusco in #98
• Keep kv-cache to the actual data stays on disk but mmap is removed af… by @tjake in #100

New Contributors

• @eltociear made their first contribution in #90
• @LuccaPrado made their first contribution in #96
• @mariofusco made their first contribution in #98

Full Changelog: v0.7.0...v0.8.0

v0.7.0

What's Changed

• Rest http port fix by @tjake in #76
• Overloads download method to specify progress by @lordofthejars in #71
• Set the max tokens based on the model and fix temp for now by @tjake in #77
• Creates ProgressReporter interface instead of functional interface by @lordofthejars in #80
• Add Qwen2 support and fix bug with small models using I8Q4 by @tjake in #82
• Add support for gemma-2 models by @tjake in #83

Full Changelog: v0.6.0...v0.7.0

v0.6.0

What's Changed

• Adds split by layer for distributed inference by @tjake in #60
• #62 global preview feature description by @AdamBien in #63
• Add Support for Q4 Embeddings and vector cleanup by @tjake in #70

New Contributors

• @AdamBien made their first contribution in #63

Full Changelog: v0.5.0...v0.6.0