{"id":1322,"date":"2026-09-17T16:57:05","date_gmt":"2026-09-17T07:57:05","guid":{"rendered":"https:\/\/localmodelwatch.tsuchitsuchi.com\/engines-en\/"},"modified":"2026-09-19T08:09:24","modified_gmt":"2026-09-18T23:09:24","slug":"engines-en","status":"publish","type":"page","link":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/engines-en\/","title":{"rendered":"Local LLM Engines and Tools: Release Tracker"},"content":{"rendered":"

Inference engines, runtimes, quantization tools and UIs for running open-weight models locally, as tracked by Local Model Watch. Each page carries the full release history compiled from GitHub, basic project facts and our articles. Version numbers and dates are collected by code, not written by an AI model.<\/p>\n

Last updated 2026-09-19 (JST). 24 projects.<\/em><\/p>\n

Inference engines and runtimes<\/h2>\n
\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Project<\/th>\nWhat it is<\/th>\nLatest release<\/th>\nReleased<\/th>\n<\/tr>\n<\/thead>\n
llama.cpp<\/a><\/td>\nLLM inference engine written in C\/C++<\/td>\nv0.4.1<\/a><\/td>\n2026-09-15<\/td>\n<\/tr>\n
Ollama<\/a><\/td>\nLocal LLM runtime that pulls and runs models with a single command<\/td>\nv0.34.2<\/a><\/td>\n2026-09-16<\/td>\n<\/tr>\n
vLLM<\/a><\/td>\nHigh-throughput LLM serving engine built around PagedAttention<\/td>\nv0.29.0<\/a><\/td>\n2026-09-09<\/td>\n<\/tr>\n
SGLang<\/a><\/td>\nLLM serving framework known for RadixAttention prefix caching and structured output<\/td>\nv0.5.20<\/a><\/td>\n2026-09-19<\/td>\n<\/tr>\n
KoboldCpp<\/a><\/td>\nSingle-file runtime that bundles llama.cpp with a web UI and API<\/td>\nv1.121<\/a><\/td>\n2026-09-16<\/td>\n<\/tr>\n
llamafile<\/a><\/td>\nPackages model weights and llama.cpp into one executable that runs on any OS without installation<\/td>\n0.10.6<\/a><\/td>\n2026-09-16<\/td>\n<\/tr>\n
LocalAI<\/a><\/td>\nSelf-hosted, OpenAI-API-compatible inference server that fronts multiple backends for text, image and audio<\/td>\nv4.10.0<\/a><\/td>\n2026-09-18<\/td>\n<\/tr>\n
MLC LLM<\/a><\/td>\nTVM-based compiler that deploys LLMs natively to phones, browsers (WebGPU) and a wide range of GPUs<\/td>\n\u2014<\/td>\n\u2014<\/td>\n<\/tr>\n
RamaLama<\/a><\/td>\nRuns local LLMs inside OCI containers, treating models as container images on Podman or Docker<\/td>\nv0.24.0<\/a><\/td>\n2026-08-22<\/td>\n<\/tr>\n
exo<\/a><\/td>\nFramework that clusters everyday devices (Macs, PCs, phones) to run a single model across them<\/td>\nv1.0.71<\/a><\/td>\n2026-04-24<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

Quantization, model formats and fine-tuning<\/h2>\n
\n\n\n\n\n\n\n\n\n
Project<\/th>\nWhat it is<\/th>\nLatest release<\/th>\nReleased<\/th>\n<\/tr>\n<\/thead>\n
ggml<\/a><\/td>\nTensor library underlying llama.cpp and whisper.cpp<\/td>\nv0.24.0<\/a><\/td>\n2026-09-14<\/td>\n<\/tr>\n
ExLlamaV3<\/a><\/td>\nInference library tuned for consumer NVIDIA GPUs<\/td>\nv1.5.0<\/a><\/td>\n2026-09-13<\/td>\n<\/tr>\n
ik_llama.cpp<\/a><\/td>\nFork of llama.cpp focused on CPU and hybrid performance, with its own quantization types (IQ*_K and others)<\/td>\n\u2014<\/td>\n\u2014<\/td>\n<\/tr>\n
Unsloth<\/a><\/td>\nLibrary for fast, low-VRAM fine-tuning (LoRA \/ QLoRA)<\/td>\nv0.1.811-beta<\/a><\/td>\n2026-09-19<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

UIs and desktop apps<\/h2>\n
\n\n\n\n\n\n\n\n\n
Project<\/th>\nWhat it is<\/th>\nLatest release<\/th>\nReleased<\/th>\n<\/tr>\n<\/thead>\n
Open WebUI<\/a><\/td>\nSelf-hosted chat UI for Ollama and OpenAI-compatible APIs, with RAG, multi-user support and plugins<\/td>\nv0.11.3<\/a><\/td>\n2026-08-31<\/td>\n<\/tr>\n
Jan<\/a><\/td>\nOffline desktop chat app with llama.cpp built in; doubles as a local API server<\/td>\nv0.8.4<\/a><\/td>\n2026-07-23<\/td>\n<\/tr>\n
text-generation-webui<\/a><\/td>\nGradio-based web UI for LLMs that switches between backends such as llama.cpp, ExLlama and Transformers<\/td>\nv4.9<\/a><\/td>\n2026-05-21<\/td>\n<\/tr>\n
LM Studio CLI (lms)<\/a><\/td>\nCommand-line tool for the LM Studio desktop app: pull, load and serve models from scripts<\/td>\n\u2014<\/td>\n\u2014<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

Image, video and audio<\/h2>\n
\n\n\n\n\n\n\n\n\n\n\n
Project<\/th>\nWhat it is<\/th>\nLatest release<\/th>\nReleased<\/th>\n<\/tr>\n<\/thead>\n
ComfyUI<\/a><\/td>\nNode-based UI for building image, video and audio generation pipelines<\/td>\nv0.36.0<\/a><\/td>\n2026-09-16<\/td>\n<\/tr>\n
Diffusers<\/a><\/td>\nHugging Face’s Python library for diffusion models; the standard way to run image, video and audio generators from code<\/td>\nv0.40.0<\/a><\/td>\n2026-08-20<\/td>\n<\/tr>\n
InvokeAI<\/a><\/td>\nDesktop\/web app for image generation with a unified canvas and workflow editor<\/td>\nv6.14.1<\/a><\/td>\n2026-09-07<\/td>\n<\/tr>\n
whisper.cpp<\/a><\/td>\nC\/C++ port of OpenAI’s Whisper speech recognition on ggml<\/td>\nv1.9.4<\/a><\/td>\n2026-09-11<\/td>\n<\/tr>\n
TTS WebUI<\/a><\/td>\nWeb UI that brings many text-to-speech and music generation models under one roof<\/td>\nv1.5.2<\/a><\/td>\n2026-09-01<\/td>\n<\/tr>\n
Stable Diffusion WebUI (AUTOMATIC1111)<\/a><\/td>\nThe classic web UI for Stable Diffusion, with a large extension ecosystem<\/td>\nv1.10.1<\/a><\/td>\n2025-02-09<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

Inference engines, runtimes, quantization tools and UIs for running open-weight models locally, as tracked by Local Model Watch. Each page carries the full release history compiled from GitHub, basic project facts and our articles. Version numbers and dates are collected by code, not written by an AI model. Last updated 2026-09-19 (JST). 24 projects. Inference […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-1322","page","type-page","status-publish","hentry"],"lang":"en","translations":{"en":1322,"ja":1320},"_links":{"self":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/pages\/1322","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/comments?post=1322"}],"version-history":[{"count":5,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/pages\/1322\/revisions"}],"predecessor-version":[{"id":1852,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/pages\/1322\/revisions\/1852"}],"wp:attachment":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media?parent=1322"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}