OpenGPTEvidence-based model decisions
Help
MODEL SELECTOR

Choose a model

Select a task, deployment, priority, and language.

1. Primary task
2. Deployment
3. Top priority
4. Language need

Best-fit families: Gemini: High match, matches 3/3; Current versions: gemini-3.5-flash, gemini-3.1-flash-lite · GPT: High match, matches 3/3; Current versions: gpt-5.6-sol, gpt-5.6-terra · Grok: High match, matches 3/3; Current versions: grok-4.5, grok-4.3

Best-fit families

Gemini

High match · matches 3/3
Why it fits
Native multimodal options across several input types.
Watch for
Closed weights for flagship hosted models.

Current versions

  • gemini-3.5-flashHighest capability
  • gemini-3.1-flash-liteSpeed and volume
Compare these versions →Official basis ↗

GPT

High match · matches 3/3
Why it fits
Covers general, coding, vision, and tool-based workflows.
Watch for
Closed weights and provider-controlled deployment.

Current versions

  • gpt-5.6-solHighest capability
  • gpt-5.6-terraQuality and cost balance
Compare these versions →Official basis ↗

Grok

High match · matches 3/3
Why it fits
Managed models include tool and real-time information paths.
Watch for
Closed weights and provider-controlled data/tool access.

Current versions

  • grok-4.5Highest capability
  • grok-4.3Quality and cost balance
Compare these versions →Official basis ↗

Family match only. Check version availability, then compare.