gpt-oss
A collection of reference inference implementations for gpt-oss by OpenAI
Decision gist · record as of 2026-08-14
Yes, with conditions. Install if you need reference code for running gpt-oss models locally and have the required GPU hardware. Verify the license terms before production use. The aging maintenance status and unavailable repository metadata mean you should confirm that the package still works with current versions of dependencies before committing to it for critical workloads.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >=3.12.
- Running inference requires significant GPU memory.
- Models must be used with the harmony response format or they will not work correctly.
License · maintenance · safety
(unclear) — License treatment is unclear; no SPDX identifier or raw license text is available in the package metadata. Verify the actual license terms before deploying in production or modifying the code.
last release 2026-01-13 (213 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 88,119 downloads/mo, #13,747 on PyPI
Alternatives
Verify before relying
pip install gpt-oss
from openai_harmony import (
load_harmony_encoding,
Conversation,
Message,
Role,
SystemContent,
)
encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)
convo = Conversation.from_messages([Message.from_role_and_content(Role.USER, "What is the weather?")])- Whether repository is actively maintained or archived—metadata unavailable
- Exact license terms and any commercial use restrictions
- GPU memory requirements for different model sizes and quantization settings
- Production readiness of PyTorch/Triton/Metal reference implementations
What it is and what it does
gpt-oss is a Python package providing reference implementations for inference with OpenAI's open-weight models: gpt-oss-120b and gpt-oss-20b. Both models use the harmony response format and support reasoning effort configuration, chain-of-thought output, function calling, and structured outputs. The package includes PyTorch, Triton, and Metal backends for different hardware targets and integrates with inference frameworks for OpenAI-compatible API serving.
The package depends on openai-harmony for format handling, plus a broad stack of web, async, and serialization libraries (fastapi, aiohttp, pydantic, lxml, requests, structlog, tenacity, uvicorn, docker, jupyter-client, tiktoken, termcolor, chz). It is designed for developers who want to run these models locally or in custom inference pipelines, with particular emphasis on agentic use cases and fine-tuning.
Use it for
- Run gpt-oss-20b locally for low-latency reasoning tasks with constrained hardware
- Deploy gpt-oss-120b for production general-purpose reasoning on high-end GPUs
- Build agentic systems using the models' native function calling and web-browsing capabilities
- Fine-tune gpt-oss models to specialized domains via parameter adjustment
- Integrate inference into fastapi applications using the reference implementations
- Debug model reasoning by accessing full chain-of-thought output
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, with conditions.
Install if you need reference code for running gpt-oss models locally and have the required GPU hardware. Verify the license terms before production use. The aging maintenance status and unavailable repository metadata mean you should confirm that the package still works with current versions of dependencies before committing to it for critical workloads.
Install
gpt-oss on PyPI
Before you install
Low install friction with a pure-Python wheel. Maintenance status is aging—the package has not been updated since its initial release, and repository metadata is unavailable, making it unclear whether active development continues.
Requires Python >=3.12. Running inference requires significant GPU memory. Models must be used with the harmony response format or they will not work correctly.
License in practice
License treatment is unclear; no SPDX identifier or raw license text is available in the package metadata. Verify the actual license terms before deploying in production or modifying the code.
Quickstart
pip install gpt-oss
from openai_harmony import (
load_harmony_encoding,
Conversation,
Message,
Role,
SystemContent,
)
encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)
convo = Conversation.from_messages([Message.from_role_and_content(Role.USER, "What is the weather?")])Verify before relying
- Whether repository is actively maintained or archived—metadata unavailable
- Exact license terms and any commercial use restrictions
- GPU memory requirements for different model sizes and quantization settings
- Production readiness of PyTorch/Triton/Metal reference implementations
Package facts
| License | Not declared unclear |
| Python support | Supports the current Python release >=3.12 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 15 packagesopenai-harmonytiktokenaiohttpchzdockerfastapihtml2textlxmlpydanticstructlogtenacityuvicornrequeststermcolorjupyter-client |
| Maintenance | Aging 213 days since the last release |
| First released | |
| Downloads | 88,119 / month, #13,747 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: gpt_oss-0.0.9-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “gpt-oss inference implementation”
- gpt-ossProvides reference implementations for running OpenAI's gpt-oss-120b…
- openai-harmonyRenders and parses the Harmony response format used by OpenAI's…
- ollamaProvides a Python client to interact with Ollama, a local or…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also openai-harmony · tpu-inference · llama-index-agent-openai · ipex-llm · vllm-tpu · liger-kernel · nvidia-modelopt · torchao · pytorch-pretrained-bert · openvino