Skip to content

Release 1.0.0 #139

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
stratika merged 1 commit into from
Jul 28, 2026
Merged

Release 1.0.0 #139

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 36 additions & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,42 @@

All notable changes to GPULlama3.java will be documented in this file.

## [1.0.0] - 2026-07-28

### Features

- Auto-detect TornadoVM backend in llama-tornado / llamaTornado ([#137](https://github.com/beehive-lab/GPULlama3.java/pull/137))
- Add TornadoVM CUDA backend support and tensor-core (MMA) accelerated batch prefill (FP16 & Q8_0, Llama & Qwen3) ([#127](https://github.com/beehive-lab/GPULlama3.java/pull/127))
- Add prefill-decode and batch-prefill-decode for Qwen3 (FP16 and Q8_0) ([#122](https://github.com/beehive-lab/GPULlama3.java/pull/122))
- Refactor GPU backend planner ([#117](https://github.com/beehive-lab/GPULlama3.java/pull/117))
- Several fixes and improvements for CI ([#115](https://github.com/beehive-lab/GPULlama3.java/pull/115))
- Ci/metrics history ([#114](https://github.com/beehive-lab/GPULlama3.java/pull/114))
- Improve collection of performance/throughput metrics ([#113](https://github.com/beehive-lab/GPULlama3.java/pull/113))
- Update TornadoVM dependency for jdk21 and fixed suffix regarding future releases ([#111](https://github.com/beehive-lab/GPULlama3.java/pull/111))
- Add Prefill–Decode Separation with Batched Prompt Ingestion and Logits Skipping ([#102](https://github.com/beehive-lab/GPULlama3.java/pull/102))

### Other Changes

- Fix IllegalArgumentException (negative ArrayList capacity) on prompts longer than max-tokens ([#128](https://github.com/beehive-lab/GPULlama3.java/pull/128))
- On-device greedy sampling (-Dllama.deviceSample): GPU argmax, keep logits device-side (~500x less D2H) ([#134](https://github.com/beehive-lab/GPULlama3.java/pull/134))
- Fix Qwen3 FP16 garbage on CUDA backend: cross-workgroup race in RMS-norm reduction ([#132](https://github.com/beehive-lab/GPULlama3.java/pull/132))
- Docs: modernize README — position as JVM-native LLM inference & serving engine ([#136](https://github.com/beehive-lab/GPULlama3.java/pull/136))
- Add llama-bench-style benchmark: llama-tornado --bench (pp/tg/pg matrix, avg±stddev t/s, md/csv/json) ([#133](https://github.com/beehive-lab/GPULlama3.java/pull/133))
- OpenAI-compatible server (llama-tornado --server): /v1/chat/completions + /v1/completions, streaming, zero-dep ([#135](https://github.com/beehive-lab/GPULlama3.java/pull/135))
- Release 0.5.0 ([#125](https://github.com/beehive-lab/GPULlama3.java/pull/125))
- Qwen3 decode: split-KV attention + backend-aware warp GEMV (FP16 & Q8_0) ([#123](https://github.com/beehive-lab/GPULlama3.java/pull/123))
- Introduce tool calling support ([#116](https://github.com/beehive-lab/GPULlama3.java/pull/116))
- Cleanup of presentation materials ([#121](https://github.com/beehive-lab/GPULlama3.java/pull/121))
- Add Q4_K/Q5_K/Q6_K GPU support via Q8_0 dequantization ([#108](https://github.com/beehive-lab/GPULlama3.java/pull/108))
- llama-tornado script curation ([#112](https://github.com/beehive-lab/GPULlama3.java/pull/112))
- Add Apple Metal backend support ([#103](https://github.com/beehive-lab/GPULlama3.java/pull/103))
- Add DevoxxGreece presentation material ([#109](https://github.com/beehive-lab/GPULlama3.java/pull/109))
- Devstral 2 support (Mistral 3 architecture, Tekken tokenizer, YaRN … ([#107](https://github.com/beehive-lab/GPULlama3.java/pull/107))
- Add llamaTornado Java 25 single-file launcher with Metal backend support ([#105](https://github.com/beehive-lab/GPULlama3.java/pull/105))
- [refactor] Simplify and unify the TornadoVM layer planner infrastructure ([#101](https://github.com/beehive-lab/GPULlama3.java/pull/101))
- AddCI Actions for Quarkus-LangChain4j integration ([#89](https://github.com/beehive-lab/GPULlama3.java/pull/89))
- Simplify and generalize TornadoVM version across JDK profiles in pom.xml ([#99](https://github.com/beehive-lab/GPULlama3.java/pull/99))

## [0.4.0] - 2026-02-25

### Other Changes
Expand Down
4 changes: 2 additions & 2 deletions CITATION.cff
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,6 @@ authors:
given-names: "Christos"
title: "GPULlama3.java"
license: MIT License
version: 0.4.0
date-released: 2026-02-25
version: 1.0.0
date-released: 2026-07-28
url: "https://github.com/beehive-lab/GPULlama3.java"
2 changes: 1 addition & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,7 @@ GPULlama3ChatModel model = GPULlama3ChatModel.builder()
<dependency>
<groupId>io.github.beehive-lab</groupId>
<artifactId>gpu-llama3</artifactId>
<version>0.4.0</version>
<version>1.0.0</version>
</dependency>

<!-- JDK 25 → use version 0.4.0-jdk25 -->
Expand Down
2 changes: 1 addition & 1 deletion pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@

<properties>
<!-- CI-friendly version: resolved by flatten-maven-plugin at build time -->
<revision>0.4.0</revision>
<revision>1.0.0</revision>
<tornadovm.base.version>5.0.0</tornadovm.base.version>
<jdk.version.suffix>-jdk21</jdk.version.suffix>
<!-- CUDA backend is only available after 5.0.0 TornadoVM version -->
Expand Down
Loading