Models for AI apps: tools and structured outputs
Shipping an AI app is more than chat. You need models that call tools, honor schemas, and survive production. This list balances capability, price, and practical tool calling / structured outputs via Geek Hub.
Models in this collection
| Model | Type | Context | Price |
|---|---|---|---|
 GPT-5.4 miniOpenAI | Chat | 272k tokens | $0.8175/M in · $4.9053/M out |
 Claude Sonnet 5Anthropic | Chat | 1M tokens | $3.2702/M in · $16.3509/M out |
 Gemini 2.5 FlashGoogle | Chat | 1M tokens | $0.327/M in · $2.7251/M out |
| GPT-4.1 mini (legacy)OpenAI | Chat | 1M tokens | $0.436/M in · $1.7441/M out |
 Mistral LargeMistral | Chat | 128k tokens | $2.1801/M in · $6.5404/M out |
| Claude Haiku 4.5Anthropic | Chat | 200k tokens | $1.0901/M in · $5.4503/M out |
| Gemini 2.5 ProGoogle | Chat | 2M tokens | $1.3626/M in · $10.9006/M out |
| GPT-5.5OpenAI | Chat | 272k tokens | $5.4503/M in · $32.7018/M out |
Why these models
We prioritized models used in real apps: mini/Flash for the hot path, Sonnet/Pro/GPT-5 for orchestration and reasoning. We skip oddballs without reliable tool use.
Use them with Geek Hub
One OpenAI-compatible base URL and API key. Swap any model id from this list without rewriting your client.
Get an API keyFAQ
- Do all of these support tool calling?
- Most modern chat models in the catalog do. Check caps on the model page; in practice GPT, Claude, Gemini, and Mistral Large are safe bets.
- How do I choose mini vs flagship in my app?
- Hot path (classify, extract, router) → mini/Flash/Haiku. Planning and critical answers → Sonnet/Pro/GPT-5.x.
- Can I use the same gateway for staging and prod?
- Yes. Separate API keys and model ids per environment; Geek Hub’s base URL stays the same.