Ollama

Verified ✅ Truly Free

The standard for local AI. Run Qwen3.8, Gemma 4, GPT-OSS, Nemotron 3.5, Flint, LFM3 and hundreds of other open models on Mac, Linux or Windows with one command, with an OpenAI-compatible and an Anthropic-compatible local API (so Claude Code works with `ollama launch claude`). Version 0.40 (late 2026) uses Apple's MLX by default on Apple Silicon. Complete privacy, zero cost, offline use. Ollama also has an optional hosted service (see Ollama Cloud).

Local AI Privacy Offline Mac/Linux/Win CLI
Get API Key Suggest Edit
4100

Overview

Provider Type

Local

API Endpoint

http://localhost:11434

Free Tier Highlights

Hardware limited

Why Choose Ollama?

Ollama is the standard way to run open models locally: one command downloads and runs Qwen3.8, Gemma 4, GPT-OSS, Nemotron 3.5 and hundreds more, with OpenAI- and Anthropic-compatible local APIs (ollama launch claude starts Claude Code on a local model).

Best for: private, offline use and local development with zero cost.

Watch out: quality and speed depend on your hardware; a 7B–30B model is the realistic range on most laptops.

Quick Start Guide

1

Download from https://ollama.com/

2

Install Application

3

Open Terminal

4

Run `ollama run llama3.1`

5

Start chatting instantly

Available Models

18 free models
Model Name ID Context Capabilities
Gemma 4 (E4B / 12B / 26B / 31B) Free
gemma4
256 000 Context
-
GPT-OSS 20B Free
gpt-oss:20b
128 000 Context
-
GPT-OSS 120B Free
gpt-oss:120b
128 000 Context
-
Nemotron 3.5 Super 120B Free
nemotron-3.5-super:120b
1 000 000 Context
-
Flint 40B (coding MoE) Free
flint:40b
256 000 Context
-
Nemotron 3.5 Bolt E4B (vision) Free
nemotron-3.5-bolt:e4b
128 000 Context
-
LFM3 1.7B Free
lfm3:1.7b
256 000 Context
-
MiMo V2.6 9B Free
mimo-v2.6:9b
128 000 Context
-
Ministral 3 8B Free
ministral-3:8b
- -
Bonsai 2 9B Free
bonsai-2:9b
- -
Qwen3.6 35B Free
qwen3.6:35b
- -
Qwen3.8 27B Free
qwen3.8:27b
- -
Gemma 4 12B Free
gemma4:12b
- -
Llama 3.2 3B Free
llama3.2:3b
128 000 tokens
-
Gemma 2 9B Free
gemma2:9b
8 000 tokens
Reasoning
Mistral Nemo 12B Free
mistral-nemo:12b
32 000 tokens
Multilingual
Phi-3.5 Mini Free
phi3.5:mini
128 000 tokens
Reasoning
DeepSeek Coder V2 Free
deepseek-coder-v2
64 000 tokens
-

Integration Examples

Ready-to-use code snippets for your applications.

main.py
from openai import OpenAI

# Ollama runs a local OpenAI-compatible server
client = OpenAI(
    api_key="ollama",
    base_url="http://localhost:11434/v1"
)

response = client.chat.completions.create(
    model="llama3.2:3b",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

Free Tier Pricing & Limits

Rate Limit

Requests per minute

Hardware limited

Daily Quota

Requests per day

Unlimited

Token Limit

Tokens per minute

Unlimited

Monthly Quota

Per month limit

Free

Use Cases

Private Document Analysis

Offline Coding Assistant

Personal Medical data processing

Embedded Systems AI

Learning & Experimentation

Secure Enterprise Deployments

Limitations & Considerations

Depends on your RAM/GPU

Laptop fans will spin up

Large models (70B+) need heavy hardware

No cloud syncing

Community Hub

Live

Join the discussion, share tips, and rate Ollama.

Quick Reactions

Add Discussion

Comments are moderated. Be helpful and respectful.

Recent Activity

0 comments

Ready to Get Started?

Join thousands of developers using Ollama

Start Building Now