Skip to main content
AdOpenFree logoPromote your productReach more potential users and drive product growth and revenue.Advertise

Models & Inference

Open models, embeddings, runtimes, and inference servers.

Favicon of ollama

ollama

Free ListingStars: 181.7K

Start building with open models.

Local Model Runners

Ollama lets you get up and running with open models such as Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, and Gemma. It offers a CLI, REST API, and libraries for Python and JavaScript to run and manage models locally.

Favicon of llama.cpp

llama.cpp

Free ListingStars: 129.5K

LLM inference in C/C++

Language Models

LLM inference in C/C++ with minimal setup and state-of-the-art performance on a wide range of hardware, locally and in the cloud.

Favicon of DeepSeek-V3

DeepSeek-V3

Free ListingStars: 104.5K

Mixture-of-Experts language model with 671B parameters

Language Models

DeepSeek-V3 is a Mixture-of-Experts language model with 671B total parameters and 37B activated per token. It supports a 128K context and is available in Base and Chat variants with open weights for local deployment and API use.

Favicon of vLLM

vLLM

Free ListingStars: 94.2K

Easy, fast, and cost-efficient LLM serving for everyone.

Inference Servers & APIs

vLLM is a high-throughput, memory-efficient inference and serving engine for LLMs. It aims to make LLM serving easy, fast, and cost-efficient for everyone.

Favicon of nanochat

nanochat

Free ListingStars: 60.1K

Minimal experimental harness for training LLMs

Language Models

nanochat is a minimal, hackable harness for training LLMs end to end on a single GPU node, covering tokenization, pretraining, finetuning, evaluation, and inference. It is pitched as the best ChatGPT that $100 can buy.

Favicon of Colibri

Colibri

Free ListingStars: 46.7K

Tiny engine, immense model.

Local Model Runners

Colibri is an open-source hierarchical inference engine that adapts LLM inference to your machine, coordinating CPU, GPU, RAM and storage while streaming mixture-of-experts weights from disk.

Favicon of Heretic

Heretic

Free ListingStars: 38.2K

Training-free language model modification

Language Models

Heretic is a training-free language model modification tool. It uses mathematical techniques and configurable scorers to steer model behavior, including removing censorship, suppressing refusals, reducing slop or sycophancy, and more.

Favicon of SGLang

SGLang

Free ListingStars: 37.2K

High-performance LLM and multimodal model serving

Multimodal Models

SGLang is a high-performance serving framework for large language and multimodal models, designed for fast, scalable inference across diverse models and hardware.

Favicon of Laya

Laya

Free ListingStars: 34K

Typed decisions with calibrated probabilities in one pass

Language Models

Laya is a non-autoregressive System 1 decision engine for typed choice, score, and yes/no questions over text. It returns calibrated probabilities in one forward pass across 100+ languages and routes requests to the right checkpoint.

Favicon of Qwen3

Qwen3

Free ListingStars: 27.7K

Large language model series by the Qwen team, Alibaba Cloud

Language Models

Qwen3 is the large language model series from Alibaba Cloud's Qwen team. It offers dense and Mixture-of-Experts models with thinking and non-thinking modes, multilingual, long-context, reasoning, coding, and agent capabilities.

Favicon of DwarfStar

DwarfStar

Free ListingStars: 24K

Local inference engine for DeepSeek, GLM, and Qwen

Inference Servers & APIs

DwarfStar is a small native inference engine for running selected DeepSeek, GLM, and Qwen models on consumer hardware with Metal, CUDA, and ROCm support. It uses the project's own GGUF files and is not a general GGUF runner.

Favicon of Surya

Surya

Free ListingStars: 21.6K

OCR, layout analysis, reading order, and table recognition

Multimodal Models

Surya is a 689M-parameter OCR model from Datalab for document intelligence, covering OCR, layout analysis, reading order, and table recognition in 90+ languages.

Favicon of Qwen3-VL

Qwen3-VL

Free ListingStars: 20K

Multimodal large language model series by Qwen team

Language Models

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud, offering dense and MoE architectures, advanced visual reasoning, long context, and expanded OCR.

Favicon of Edge0

Edge0

Free ListingStars: 11K

Device-native foundation models for on-device AI

Language Models

Edge0 provides device-native foundation models and an open-source streaming MoE inference framework, running large sparse models such as 145B on a laptop and 35B on a phone with SSD-streamed weights, offline on-device processing, and zero

Favicon of kimi-k3-in-c

kimi-k3-in-c

Free ListingStars: 10K

2.78T-parameter Kimi K3 inference in portable C99

Language Models

Runs Kimi K3, a 2.78-trillion-parameter model, on a single CPU in 8.24 GB of RAM by streaming the model from disk instead of holding it in memory. Portable C99 with no BLAS, framework, or GPU.

Favicon of Kev

Kev

Free ListingStars: 9.6K

Small Jev-like decision models to train and run yourself.

Language Models

Kev is a family of small decision models built on Qwen3.5 and Qwen3.8, based on Jev's architecture. Use pretrained weights or train your own, with yes/no, multiple-choice, and rating questions in one request.

Favicon of DeepInfer

DeepInfer

Free ListingStars: 7.9K

High-performance inference framework for DeepSeek models

Inference Servers & APIs

DeepInfer is a high-performance, large-scale inference framework for DeepSeek models. This open-source release provides offline inference with optimized kernels, parallelism, scheduling, deterministic sampling, and structured outputs.

Favicon of Gemini-API

Gemini-API

Free ListingStars: 3.6K

Async Python wrapper for Google Gemini web app

Language Models

An asynchronous Python wrapper for the Google Gemini web app (formerly Bard), offering text, image, video, audio, chat, streaming, and deep research capabilities through a reverse-engineered client.

Favicon of neuron-explorer

neuron-explorer

Free ListingStars: 1.3K

Steering-first neuron explorer for Qwen3.5-4B

Language Models

A steering-first explorer for Qwen3.5-4B that pushes every MLP neuron, keeps every rollout, and lays the results out over a 3D view of the model.

Favicon of openTPU

openTPU

Free ListingStars: 0.9K

Open-source AI accelerator: RTL, ISA, simulator, compiler

Quantization & Acceleration

openTPU is an open-source AI accelerator developed by AI, combining RTL, ISA, simulator, compiler, and profiler in one repository. It runs Qwen3, LFM2.5, and Qwen3.5 on a Kintex-7 PCIe card.

Ad
OpenFree logoPromote your product

Reach more potential users and drive product growth and revenue.

Advertise