OpenAI model

GPT-5 Mini API

Send openai/gpt-5-mini when you want 400k OpenAI vision and tools without GPT-5 prices. Listed cost is 0.25 / 2. We measured this exact family id at 80% Safe Hit Rate and 0% Bad Hit Rate on 50 prompts. Skip it if you need GPT-5 Pro quality spend, or if 8 of 10 exact cache replays is not enough.

Read quickstartEstimate cost
OpenAIPlatform curatedCatalog
ThinkingFunction CallingVision
Input price$0.25 / 1M tokens
Output price$2 / 1M tokens
Context length400k tokens
Max output8.2k tokens

What is GPT-5 Mini API in NextModel?

Send openai/gpt-5-mini when you want 400k OpenAI vision and tools without GPT-5 prices. Listed cost is 0.25 / 2. We measured this exact family id at 80% Safe Hit Rate and 0% Bad Hit Rate on 50 prompts. Skip it if you need GPT-5 Pro quality spend, or if 8 of 10 exact cache replays is not enough.

GPT-5 Mini via NextModel: 400k-token context; $0.25 / 1M tokens in / $2 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.

Best use cases

  • High-volume OpenAI calls that still need a 400,000 token window.
  • Workloads that benefit from exact-cache replay; this is the highest Safe Hit Rate in the published table.
  • A cheaper default than openai/gpt-5 at 1.25 / 10, same listed window.

When not to use

  • You need every exact repeat to replay. 2 of 10 did not.
  • You are buying Pro. GPT-5 Pro is 15 / 120 and unmeasured on cache.
  • Nano is enough. GPT-5 Nano is 0.05 / 0.40, also unmeasured on cache.

OpenAI-compatible code example

request.sh
curl https://api.nextmodel.app/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NEXTMODEL_API_KEY" \
  -d '{"model":"openai/gpt-5-mini","messages":[{"role":"user","content":"Reply with OK."}]}'

Switching models only changes the model field.

Try it in the playground

Similar alternatives

OpenAICatalog

GPT-4.1

GPT-4.1 via NextModel: 1M-token context; $2 / 1M tokens in / $8 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.

$2 / 1M tokensInput$8 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
View details
OpenAICatalog

GPT-4.1 Mini

GPT-4.1 Mini via NextModel: 1M-token context; $0.4 / 1M tokens in / $1.60 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.

$0.4 / 1M tokensInput$1.60 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
View details
OpenAICatalog

GPT-4.1 Nano

GPT-4.1 Nano via NextModel: 1M-token context; $0.1 / 1M tokens in / $0.4 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.

$0.1 / 1M tokensInput$0.4 / 1M tokensOutput1MContext
Best forimage understanding, multimodal chat
StreamingTool callingJSON modeVision
Platform curatedCurated and maintained in the NextModel catalog
View details

Compare GPT-5 Mini

GPT-5 Mini vs GPT-4.1GPT-5 Mini vs GPT-4.1 MiniGPT-5 Mini vs GPT-4.1 NanoGPT-5 Mini vs GPT-4oGPT-5 Mini vs GPT-4o-mini

Reading

Articles that explain GPT-5 Mini

FAQ

GPT-5 Mini API questions

What model id do I send for GPT-5 Mini?

Send openai/gpt-5-mini to https://api.nextmodel.app/v1.

What did CacheSafety show?

50 prompts, 50 valid responses. Safe Hit Rate 80%, Bad Hit Rate 0%, trap failure 0%. Exact-cache semantics only.

How does Mini compare to GPT-5 on price?

Mini is 0.25 / 2. GPT-5 is 1.25 / 10. Both list 400,000 tokens.

Start now

Ready to start building?

Passwordless signup, pay-as-you-go billing, and one OpenAI-compatible endpoint for every model. Create your key and send your first request in minutes.

Get your API keyRead the quickstart