jishie

GPT-4.1 mini

cloudflare-ai-gateway/openai/gpt-4.1-mini · family gpt-mini · in: text, image, pdf → out: text · price provider-declared (via models.dev). Capabilities: toolstextimagepdf.

$0.70Blended $/1M (3:1)
$0.4Input $/1M
$1.6Output $/1M
$2.001M in + 1M out
$0.1 −75%Cache read $/1M
#2666 of 5,863Cheapest rank
Provider
Cloudflare AI Gateway (cloudflare-ai-gateway)
Kind
text (outputs text)
Context
1,047,576 tokens
Max output
32,768 tokens
Capabilities
toolstextimagepdf

Same model, other providers

“GPT-4.1 mini” is offered by 18 providers — cheapest-effective first. This one (Cloudflare AI Gateway) is the cheapest.

ProviderInputOutputBlendedContext
Cloudflare AI Gateway (this)$0.4$1.6$0.701M
OrcaRouter$0.4$1.6$0.701M
Impossibl$0.4$1.6$0.701M
OpenRouter$0.4$1.6$0.701M
OpenAI$0.4$1.6$0.701M
NEAR AI Cloud$0.4$1.6$0.701M
Ofox$0.4$1.6$0.701M
Azure Cognitive Services$0.4$1.6$0.701M
Azure$0.4$1.6$0.701M
Vercel AI Gateway$0.4$1.6$0.701M
LLM Gateway$0.4$1.6$0.701M
Kilo Gateway$0.4$1.6$0.701M
Pioneer$0.4$1.6$0.701M

Call it

jishie indexes & prices models; it does not proxy inference.

Most providers are OpenAI-compatible — point base_url at Cloudflare AI Gateway 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-4.1-mini","messages":[{"role":"user","content":"hello"}]}'

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