gpt context compression
Compress GPT chat context to reduce tokens for later requests.
POST
/v1/responses/compactBody Parametersapplication/json
input* array
List of historical messages that need to be compressed
items object object
model* string
Model name, such as gpt-5-codex
Responses
200 application/json
response-200 object
400 Bad request or validation failed application/json
response-400 object
properties error* object
401 Missing or invalid API Key application/json
response-401 object
1{
2"input": [
3{
4"content": "Hello, this is a compact test.",
5"role": "user",
6"type": "message"
7}
8],
9"model": "gpt-5.4"
10}
1curl -X POST "https://api.digislash.com/v1/responses/compact" \
2-H "Authorization: Bearer <API Key>" \
3-H "Content-Type: application/json" \
4-d '{
5"input": [
6{
7"content": "Hello, this is a compact test.",
8"role": "user",
9"type": "message"
10}
11],
12"model": "gpt-5.4"
13}'
1▼{
2"created_at": 1779183068,
3"id": "resp_010f1725ac1c2947016a0c2ddb9dcc81989aff546cbc9f7eb2",
4"object": "response.compaction",
5▼"output": [
6▶{ 5 keys },
7▶{ 3 keys }
8],
9▼"usage": {
10"input_tokens": 31,
11▶"input_tokens_details": { 1 keys },
12"output_tokens": 34,
13▶"output_tokens_details": { 1 keys },
14"total_tokens": 65
15}
16}