jishie

GPT-5.3 Codex

impossibl/openai/gpt-5.3-codex · family gpt-codex · price provider-declared (via models.dev), USD per 1M tokens. Capabilities: toolsreasoningtextimagepdf.

$4.81Blended $/1M (3:1)
$1.75Input $/1M
$14Output $/1M
$15.751M in + 1M out
$0.175 −90%Cache read $/1M
#4856 of 5,000Cheapest rank
Provider
Impossibl (impossibl)
Context
400,000 tokens
Max output
128,000 tokens
Capabilities
toolsreasoningtextimagepdf

Same model, other providers

“GPT-5.3 Codex” is offered by 9 providers — cheapest-effective first. This one (Impossibl) is the cheapest.

ProviderInputOutputBlendedContext
Impossibl (this)$1.75$14$4.81400k
Vivgrid$1.75$14$4.81400k
FastRouter$1.75$14$4.81400k
LLM Gateway$1.75$14$4.81400k
OpenCode Zen$1.75$14$4.81400k
GitHub Copilot$1.75$14$4.81400k
Neon$1.75$14$4.81400k
FrogBot$1.75$14$4.81400k
Pioneer$1.75$14$4.81400k

Call it

jishie indexes & prices models; it does not proxy inference. Most providers are OpenAI-compatible — point base_url at Impossibl and pass this model id:

curl $BASE_URL/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-5.3-codex","messages":[{"role":"user","content":"hello"}]}'

Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):