Chat Completions

POST

https://api.reka.ai/v1/chat/completions

POST

/v1/chat/completions

$ curl -X POST https://api.reka.ai/v1/chat/completions \
>      -H "Authorization: Bearer <token>" \
>      -H "Content-Type: application/json" \
>      -d '{
>   "messages": [
>     {
>       "role": "user",
>       "content": "string"
>     }
>   ],
>   "model": "reka-flash-research"
> }'

200Successful

1 {
2   "id": "string",
3   "choices": [
4     {
5       "finish_reason": "stop",
6       "index": 1,
7       "message": {
8         "role": "assistant",
9         "reasoning_steps": [
10           {
11             "role": "assistant",
12             "reasoning_content": "string",
13             "tool_calls": [
14               {
15                 "id": "string",
16                 "name": "search_web",
17                 "args": {
18                   "urls": [
19                     "string"
20                   ],
21                   "task": "string"
22                 }
23               }
24             ],
25             "content": {
26               "tool_name": "search_web",
27               "tool_output": [
28                 {
29                   "url": "string",
30                   "title": "string",
31                   "snippet": "string",
32                   "age": "string"
33                 }
34               ],
35               "tool_error": "string"
36             },
37             "tool_call_id": "string"
38           }
39         ],
40         "reasoning_content": "string",
41         "content": "string",
42         "annotations": [
43           {
44             "type": "url_citation",
45             "url_citation": {
46               "end_index": 1,
47               "start_index": 1,
48               "url": "string",
49               "title": "string"
50             }
51           }
52         ]
53       }
54     }
55   ],
56   "created": 1,
57   "model": "string",
58   "usage": {
59     "completion_tokens": 0,
60     "prompt_tokens": 0,
61     "total_tokens": 0
62   }
63 }

curl -X POST https://api.reka.ai/v1/chat/completions \
     -H "Authorization: Bearer <token>" \
     -H "Content-Type: application/json" \
     -d '{
  "messages": [
    {
      "role": "user",
      "content": "string"
    }
  ],
  "model": "reka-flash-research"
}'

{
  "id": "string",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 1,
      "message": {
        "role": "assistant",
        "reasoning_steps": [
          {
            "role": "assistant",
            "reasoning_content": "string",
            "tool_calls": [
              {
                "id": "string",
                "name": "search_web",
                "args": {
                  "urls": [
                    "string"
                  ],
                  "task": "string"
                }
              }
            ],
            "content": {
              "tool_name": "search_web",
              "tool_output": [
                {
                  "url": "string",
                  "title": "string",
                  "snippet": "string",
                  "age": "string"
                }
              ],
              "tool_error": "string"
            },
            "tool_call_id": "string"
          }
        ],
        "reasoning_content": "string",
        "content": "string",
        "annotations": [
          {
            "type": "url_citation",
            "url_citation": {
              "end_index": 1,
              "start_index": 1,
              "url": "string",
              "title": "string"
            }
          }
        ]
      }
    }
  ],
  "created": 1,
  "model": "string",
  "usage": {
    "completion_tokens": 0,
    "prompt_tokens": 0,
    "total_tokens": 0
  }
}

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <token>, where token is your auth token.

Request

This endpoint expects an object.

messageslist of objectsRequired

A list of messages comprising the conversation so far.

modelstringRequiredDefaults to reka-flash-research

The model name used to generate the response. Currently, only "reka-flash-research" is supported for this endpoint.

streambooleanOptionalDefaults to false

Determines whether to stream the response incrementally. If set to true, the agent will return a stream of response chunks. If false, the response will be returned as a single object.

researchobjectOptional

response_formatobjectOptional

An object specifying the format that the agent must output. Setting this to { "type": "json_schema", "json_schema": {...} } enables Structured Outputs, which ensures the agent returns output that matches the supplied JSON schema. If not specified, the agent will return a plain text response.

Response

CreateChatCompletionResponseobject

A chat completion response from the agent, containing the generated message and other metadata. This is used when stream is set to false in the request.

CreateChatCompletionStreamResponseobject

A streamed response from the agent, containing a chunk of the chat completion. This is used when stream is set to true in the request.

$	curl -X POST https://api.reka.ai/v1/chat/completions \
>	-H "Authorization: Bearer <token>" \
>	-H "Content-Type: application/json" \
>	-d '{
>	"messages": [
>	{
>	"role": "user",
>	"content": "string"
>	}
>	],
>	"model": "reka-flash-research"
>	}'

1	{
2	"id": "string",
3	"choices": [
4	{
5	"finish_reason": "stop",
6	"index": 1,
7	"message": {
8	"role": "assistant",
9	"reasoning_steps": [
10	{
11	"role": "assistant",
12	"reasoning_content": "string",
13	"tool_calls": [
14	{
15	"id": "string",
16	"name": "search_web",
17	"args": {
18	"urls": [
19	"string"
20	],
21	"task": "string"
22	}
23	}
24	],
25	"content": {
26	"tool_name": "search_web",
27	"tool_output": [
28	{
29	"url": "string",
30	"title": "string",
31	"snippet": "string",
32	"age": "string"
33	}
34	],
35	"tool_error": "string"
36	},
37	"tool_call_id": "string"
38	}
39	],
40	"reasoning_content": "string",
41	"content": "string",
42	"annotations": [
43	{
44	"type": "url_citation",
45	"url_citation": {
46	"end_index": 1,
47	"start_index": 1,
48	"url": "string",
49	"title": "string"
50	}
51	}
52	]
53	}
54	}
55	],
56	"created": 1,
57	"model": "string",
58	"usage": {
59	"completion_tokens": 0,
60	"prompt_tokens": 0,
61	"total_tokens": 0
62	}
63	}