Skip to main content

vLLM plugins for LLM-jp models

This repository contains vLLM plugins which are necessary to launch LLM-jp models with vLLM entrypoints.

Installation

uv pip install llm-jp-vllm

The above command installs llm-jp-vllm directory onto your environment's site-packages.

Usage

This repository works with the plugin mechanism on vLLM.

After necessary change in vLLM was merged, you can use reasoning/tool parsers implemented in this repository by passing appropriate module names to --reasoning/tool-parser-plugin option.

vllm serve {llm-jp-4 model} \
  --trust-remote-code \
  --reasoning-parser llmjp4 \
  --reasoning-parser-plugin llm_jp_vllm.llmjp4

Tool calling

The llmjp4 tool parser converts Harmony-format tool calls (including the parallel call extension of LLM-jp-4) into OpenAI-compatible tool_calls, for both streaming and non-streaming responses:

vllm serve {llm-jp-4 model} \
  --trust-remote-code \
  --reasoning-parser llmjp4 \
  --reasoning-parser-plugin llm_jp_vllm.llmjp4 \
  --enable-auto-tool-choice \
  --tool-call-parser llmjp4 \
  --tool-parser-plugin llm_jp_vllm.llmjp4

Development

uv sync --all-extras --dev
./checks.sh  # ruff / mypy
uv run pytest

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

llm_jp_vllm-0.1.0.tar.gz (13.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

llm_jp_vllm-0.1.0-py3-none-any.whl (14.7 kB view details)

Uploaded Python 3

File details

Details for the file llm_jp_vllm-0.1.0.tar.gz.

File metadata

  • Download URL: llm_jp_vllm-0.1.0.tar.gz
  • Upload date:
  • Size: 13.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for llm_jp_vllm-0.1.0.tar.gz
Algorithm Hash digest
SHA256 c0f68304d3bc915f27f6efef54fb8d0407d754aacaf78a11e772b2598433d773
MD5 f0027535d6e5c77e3c7b774a85afc578
BLAKE2b-256 b103640a96eab460c1ccba66cae9656df184445e232f96e3c31242d08647c9cd

See more details on using hashes here.

Provenance

The following attestation bundles were made for llm_jp_vllm-0.1.0.tar.gz:

Publisher: release.yml on llm-jp/llm-jp-vllm

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file llm_jp_vllm-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: llm_jp_vllm-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 14.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for llm_jp_vllm-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 023e28a2df2e084199c1d4da897d8e6f2d71944ccbe00c279aaf40e4a4f09c0a
MD5 bd948f4c45c64855f398beea9fb5e042
BLAKE2b-256 0a24d9c9f854d76a015d3508d4683ad37616d01bb373ca8e7135a142c551ccd8

See more details on using hashes here.

Provenance

The following attestation bundles were made for llm_jp_vllm-0.1.0-py3-none-any.whl:

Publisher: release.yml on llm-jp/llm-jp-vllm

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page