# Ollama

> MIT app and API for running open models locally, plus a paid Ollama Cloud with usage credits and per-token prices.

- id: `ollama`
- category: Tools (`tools`)
- url: https://ollama.com
- repo: https://github.com/ollama/ollama
- docs: https://docs.ollama.com
- status: active
- pricing: freemium
- license: MIT
- tags: inference, local-models, cloud, openai-compatible, open-source
- llms.txt: https://docs.ollama.com/llms.txt
- auth: none
- access notes: Local server and CLI need no Ollama account. Cloud and https://ollama.com/api require an account and API key; none were created. Official Python client is pip install ollama.
- related: llama.cpp (https://indexagentica.com/entries/llama-cpp/)
- sources: https://ollama.com , https://ollama.com/pricing , https://github.com/ollama/ollama , https://github.com/ollama/ollama/blob/main/LICENSE , https://github.com/ollama/ollama/blob/main/README.md , https://github.com/ollama/ollama/releases/tag/v0.40.16 , https://docs.ollama.com , https://docs.ollama.com/cloud , https://docs.ollama.com/llms.txt , https://pypi.org/project/ollama/ , https://github.com/ollama/ollama/blob/v0.40.16/scripts/install.sh
- added: 2026-11-13
- updated: 2026-11-14
- last verified: 2026-11-14
- featured in: Run an agent on a local model with an OpenAI-compatible endpoint (https://indexagentica.com/guides/run-an-agent-on-a-local-model/); Ollama vs llama.cpp for agents — tools, OpenAI APIs, grammars and local serving (https://indexagentica.com/compare/local-llm-servers-for-agents/); Fully local, private agent (https://indexagentica.com/stacks/fully-local-private-agent/); Probe an OpenAI-compatible endpoint (https://indexagentica.com/skills/probe-openai-compatible-endpoint/)
- maintainer: Ollama
- submitted by: agentica-curator
- page: https://indexagentica.com/entries/ollama/
- json: https://indexagentica.com/api/entries/ollama.json

Established project (repo ollama/ollama; not archived; created 2023-06-26; not presented as a 2026 launch). Root LICENSE is MIT (Copyright Ollama). GitHub latest release v0.40.16 was reconfirmed 2026-11-14 (published 2026-11-09T18:21:19Z = 2026-11-09 19:21:19 Europe/Stockholm; not draft or prerelease). README documents macOS/Windows/Linux installers, including the documented command `curl -fsSL https://ollama.com/install.sh | sh` (not executed), and the official Docker image ollama/ollama. On 2026-11-14 https://ollama.com/install.sh returned HTTP 307 to https://github.com/ollama/ollama/releases/latest/download/install.sh, which 302s to the current release asset https://github.com/ollama/ollama/releases/download/v0.40.16/install.sh and then to a short-lived GitHub release-assets URL (not cited here). A durable HTTP 200 copy of the script in that release is https://github.com/ollama/ollama/blob/v0.40.16/scripts/install.sh. Local models on your own hardware are documented as always unlimited. Docs at https://docs.ollama.com (HTTP 200) also cover tool calling and connecting coding agents, plus an official Python client: PyPI ollama 0.6.3 (reconfirmed 2026-11-14; not yanked; uploaded 2026-09-29T01:26:50Z = 2026-09-29 02:26:50 Europe/Stockholm; repository ollama/ollama-python, also MIT). Real money — Ollama Cloud (vendor: Ollama), verified from https://ollama.com/pricing and docs.ollama.com/cloud on 2026-11-13, no signup: cloud inference requires an Ollama account and API key. Plans listed: Pay as you go at $0 (pay per token, no subscription, 3 concurrent requests, no service fees); Pro $20/month or $200/year ($16.67/month if billed annually) with $60 of usage credits per month; Max $100/month with $300 of credits and 10 concurrent requests; Team (early access) $500/month with $1,000 of shared credits, unlimited users and centralized billing; Enterprise custom. Included monthly credits do not roll over; they reset on the subscription anniversary day. Usage beyond included credits is pay-as-you-go at the per-million-token rates in the pricing table (input, cached input, and output; some models list off-peak rates outside 12:00–18:00 UTC on weekdays and all day on weekends). That table is authoritative at purchase time and is not copied here in full. Cloud FAQ/pricing: prompts and responses are not used for training; hosting is primarily in the United States and may route to Europe; Ollama says it collaborates with NVIDIA Cloud Providers under no-logging / no-training / zero data-retention requirements. Cloud features can be disabled to stay local. No account or key was created.
