Ollama
Named by 50 kept items in this workspace.
40 items name it under frameworks and tools and 10 under classification.technologies.
The items that name it 50
Runs a local LLM for the offline --local mode.
Runs a System One decision model on a local server or in Ollama's cloud; needs Ollama 0.35.0 or later, or 0.40.5 or later for cloud models.
Local model runner offering Liquid LFM3; text model fits in about 1GB at 4-bit.
Runs the Qwen model locally and serves it at http://127.0.0.1:11434/v1 for OpenCode. Started with OLLAMA_NO_CLOUD=1 to disable cloud features and OLLAMA_CONTEXT_LENGTH=65536.
Runs open models locally. Installs on macOS, Linux, Windows or as the ollama/ollama Docker image. Pulls models from its library, serves a REST API on localhost:11434 (/api/chat), and is described as having 184k GitHub stars and an MIT licence.
Local model endpoint the PRAXO-CORE quickstart can run against.
Local model provider that 5ire can connect to.
Supported model endpoint option, including self-hosted models.
Local model runner with early support for Nemotron 3.5 Bolt; Ollama Cloud offers M3.1 Flash free for a limited time.
Supported provider in AnyLanguageModel.
Runs the open-weight models locally. Its default temperature of 0.8 was used in the repeated-sampling experiment, with 100 responses sampled per unchanged request.
Runs the local GPU bge-large embedding model in the first tier of the cascade.
Run models locally, e.g. on a graphics card with 8GB of video memory or an Apple M-series Mac
Serves Reflection Beam free for a limited time, US-hosted with zero data retention.
Serving Reflection Beam free for limited time, US-hosted, zero data retention
Runs language models; cited as the comparison for Ollaya.
Local model option for Graphicle's written answers and memory proposals
Local model server that Apogee connects to over 127.0.0.1 for larger models
Local model runner with day-0 support for EmbeddingGemma 2
Used to serve the model efficiently
Platform now hosting Cloudflare's Clef decision models
Runs GGUF-quantized models for local deployment, named alongside llama.cpp as a GGUF deployment target.
Local inference platform running Tev1 0.8B classifier at ~50ms E2E
local model runner; used to run Tev1 0.8B classifier
Listed as a supported vendor.
Local model server supported as a magpie provider preset needing no key
Local AI tool powered by llama.cpp's inference engine.
Example of an OpenAI-compatible provider that can be used as the LLM upstream
Serves /v1/systemone locally since 0.35; first decision models in its library are nimble and tev1.
Local LLM backend, used for self-hosted runs and evals.
A tool the author used to try to generate text with a fixed random seed for bonus material in one of his books; he notes it wasn't always deterministic in practice.
Local serving option usable as an OpenAI-compatible endpoint for the AliangBoard agent
Running open-weight models on your own machines
Open source tool that turns any model into a local API that speaks OpenAI's language; used to pull and run local models
Mentioned as a possible OpenAI-compatible backend.
lets the engine boot in a degraded self-hosted mode with no external API key, via OLLAMA_URL pointing at a local Ollama server
Supports running local models with Sim.
Runs local models as a provider for Rikugan.
Local model deployment with streaming and multimodal capabilities
Local model runner used to pull and run the GGUF model via a single command
In every workspace you can read 121
AI
Security
All technologies in this workspace.