Skip to content
OmniRail
← All models

OpenAI · Language model

GPT Astra Latest

~openai/gpt-astra-latest
  • Language
Released Sep 11, 2026

This model always redirects to the latest model in the GPT Astra family.

Specifications

Context window
1,050,000 tokens
Max output
128,000 tokens
Input
file, image, text
Output
text
Reasoning effort
max, xhigh, high, medium, low (always on)
Tokenizer
Router
Moderation
provider-moderated

Supported parameters

Accepted by the model and forwarded as-is by the gateway.

  • max_completion_tokens
  • max_tokens
  • reasoning
  • reasoning_effort
  • response_format
  • seed
  • structured_outputs
  • tool_choice
  • tools
  • verbosity

Use it

POST /api/v1/chat/completions · MCP chat · full reference

curl https://omnirail.org/api/v1/chat/completions \
  -H "Authorization: Bearer $OMNIRAIL_KEY" -H "Content-Type: application/json" \
  -d '{"model":"~openai/gpt-astra-latest","messages":[{"role":"user","content":"Hello"}],"max_tokens":300}'