← Back | TextMate AI

Which TextMate AI Model Should You Use? A Guide to All 7 Models

2026-07-02 · Thamodharan Ganesan

TextMate AI isn't locked to a single model. Every chat lets you pick from 7 models across multiple providers, two of which are available on the Free plan.

Free-tier models

ModelBest for
Mistral Large (default)Balanced, general-purpose, largest context of the free models — the right default for most conversations. 262K context window.
GPT-OSS 20BA lightweight, fast model for quick questions where you don't need Mistral Large's full context.

Pro-only models

ModelBest for
DeepSeek V3.1Strong at reasoning and coding — reach for it on harder technical questions.
GLM 4.7Served on Cerebras for the fastest inference in the lineup — great when you want a quick answer from a different model family.
Llama 3.3 70BServed on Groq for the lowest latency in the lineup.
Nemotron 3 UltraNVIDIA's largest model here — a huge context window for feeding it long documents or conversations.
Gemma 4 31BGoogle's model, served on Cerebras for ultra-fast inference — another quick second opinion from a different model family.

Output length varies by model

Every model has a per-response output cap. Mistral Large, DeepSeek V3.1, and Nemotron 3 Ultra support up to 65,536 output tokens — enough for long rewrites or documents. GPT-OSS 20B supports up to 120,000. GLM 4.7 caps at 40,000, Llama 3.3 70B at 4,096 (Groq's free tier is the tightest of the lineup), and Gemma 4 31B uses a conservative 8,192-token cap. If you're generating long-form content in one shot, Mistral Large, DeepSeek V3.1, or Nemotron 3 Ultra are the models built for that.

How to pick

  • Default / unsure: stick with Mistral Large.
  • Need a quick, lightweight answer: GPT-OSS 20B (free).
  • Coding or hard reasoning: DeepSeek V3.1 (Pro).
  • Fastest response: GLM 4.7 or Llama 3.3 70B (Pro).
  • Very long documents/conversations: Nemotron 3 Ultra (Pro).
  • Want a fast second opinion from another model family: Gemma 4 31B (Pro).

See the Models page for the live picker with plan requirements, or Pricing for what Free vs Pro unlocks.