Context window
1.05M
OpenAI API model
OpenAI: gpt-6.1-terra
GPT-6.1 Terra for text, reasoning and function tools through streamed native Responses. This listing supports complete client-supplied conversation history.
Specs
Pricing and API details.
Transparent USD pricing. No mainland-China account or phone required. Standard model rates follow providers' international list prices at 100%; some models have a separate Paid rate after the first top-up.
Modalities
Text
Input $/1M
$2
Output $/1M
$7.5
Endpoint
OpenAI-compatible /v1/responses
About gpt-6.1-terra
GPT-6.1 Terra for text, reasoning and function tools through streamed native Responses. This listing supports complete client-supplied conversation history.
Pricing
Transparent USD pricing.
gpt-6.1-terra
$2/M input, $7.5/M output tokens. Paid in USD. Standard model rates follow providers' international list prices at 100%; some models have a separate Paid rate after the first top-up. View live pricing Above 272000 input tokens, the entire request uses the long-context tier: input and cache rates are 2x, output 1.5x. Paid input/output/cache-read rates are 55% of Standard; cache-write remains 100%. The quoted Enterprise GPT schedule is 50% for input/output/cache read and 55% for cache write, subject to an active contract; a top-up alone does not activate Enterprise.
Paid price after first top-up (input / output per 1M tokens): $1.1 / $4.125 (−45%)
Price by tier
| Tier | Standard price (input / output per 1M tokens) | Paid price after first top-up |
|---|---|---|
| Over 272K input tokens | $2 / $7.5 | $1.1 / $4.125 (−45%) |
| Up to 272K input tokens | $1 / $5 | $0.55 / $2.75 (−45%) |
Quickstart
Copy-paste API request.
curl -N https://api.chinaapi.ai/v1/responses \
-H "Authorization: Bearer $CHINAAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6.1-terra",
"stream": true,
"store": false,
"reasoning": {"effort": "none"},
"include": ["reasoning.encrypted_content"],
"input": "Reply with a short hello."
}'
Read the stream through response.completed and use its output and usage. For another turn, append every original output item and the next user message to input. For a function tool, send function_call_output with the original call_id, then submit the full input again. Use store:false; previous_response_id continuation is unavailable. Suppliers may exceed max_output_tokens, so monitor actual usage and set account quotas. Cache discounts apply only to reported cache hits; a hit is not guaranteed.
Internal links
Related model pages and guides.
How do I use gpt-6.1-terra?
Use gpt-6.1-terra through ChinaAPI with an API key and the OpenAI-compatible endpoint shown above.
How much does gpt-6.1-terra cost?
gpt-6.1-terra is $2/M input, $7.5/M output tokens, paid in USD. The live pricing page is authoritative. After the first top-up, the Paid price is $1.1/M input, $4.125/M output tokens (−45%). Standard model rates follow providers' international list prices at 100%; some models have a separate Paid rate after the first top-up.
Does gpt-6.1-terra need a Chinese account?
No. ChinaAPI provides access without a mainland-China account or phone number and includes a $2 free trial. Check live pricing for the displayed USD rate.