PocketLLM is permanently free. You only pay the AI providers you already trust, with zero markups or hidden fees.
Everything PocketLLM offers today and tomorrow — no credit card, no trial, no catches.
PocketLLM never proxies or stores provider billing. You connect directly to OpenAI, Google, Groq, or local Ollama.
No markup. No middleman. You decide which models to enable, set your own rate limits, and can rotate keys anytime inside Settings.
Read the setup guideEvery provider posts public price sheets. PocketLLM simply helps you keep track.
| Provider | Model | Input / Output | Free tier |
|---|---|---|---|
| OpenAI | GPT-4 Turbo | $10 / $30 per 1M tokens | — |
| OpenAI | GPT-3.5 | $0.50 / $1.50 per 1M tokens | — |
| OpenAI | DALL·E 3 | $0.04 – $0.12 per image | — |
| Gemini Pro | $0.50 / $1.50 per 1M tokens | ✓ Free tier | |
| Groq | Mixtral | $0.24 / $0.24 per 1M tokens | ✓ 14,400 req/day |
| Groq | Llama 3 | $0.05 / $0.08 per 1M tokens | ✓ 14,400 req/day |
| Ollama | All local models | FREE | ✓ Unlimited local |
Move the slider and pick your go-to model. PocketLLM stays free — these totals come straight from the provider price sheets.
Estimated monthly
$8
Input tokens
$2
Output tokens
$6
Assumes ~2.5k tokens per message split evenly between input/output. Adjust the slider to match your workload — 100 msgs/day on GPT-3.5 is roughly $15 per month.
PocketLLM stays free, while provider pay-as-you-go keeps costs predictable.
ChatGPT Plus
$20/mo
GPT-4 access only, single provider
Claude Pro
$20/mo
Claude family only, capped usage
PocketLLM
$0 app + pay-as-you-go
Pick any model, control spend
Download PocketLLM, add your provider keys later, and start chatting instantly. Configuration can wait until you're ready.