fix: offer a half-precision KV cache behind a flag, default off

bd3bb4e

Select commit

Loading

Failed to load commit list.

There was an error while loading. Please reload this page.

Model Integration on: pull_request

GGUF model integration

Large Model Integration on: pull_request 2

Qwen2.5-Coder 7B pure-Java slow test

DeepSeek-Coder 6.7B pure-Java slow test

Qwen3 8B pure-Java slow test

DeepSeek-R1-Distill-Qwen-7B pure-Java slow test

HuatuoGPT-o1-7B pure-Java slow test

SQLCoder-7B-2 pure-Java slow test

SmolLM3 3B pure-Java slow test

Fin-R1 7B pure-Java slow test

MFCQI on: pull_request

MFCQI score and badges

CodeQL on: pull_request 1

CodeQL Analysis

Native Kernels on: pull_request

Resolve private test version

Native linux-x86_64

Native linux-aarch64

Native macos-x86_64

Native macos-aarch64

Native windows-x86_64

Native windows-aarch64

Java backend JAR

Standalone private-test JAR

Publish private test packages

CI on: pull_request

build

Execute Java notebooks

Code scanning results

CodeQL

GGUF model integration

succeeded Aug 7, 2026 in 25m 12s

Loading