Get your prompts out of your codebase.

And save everyone on your team hours of their day

Collaborative Prompt Creation

With version control, commenting, labels, response histories and more, you can develop prompts alongside the rest of your team without all the back and forth.

Prompt version history

Testing & iteration

Know when you’re making improvements. Populate test data for your prompts, test different models, and feed production requests back into your tests.

Test data table

Deployment & monitoring

Expose your prompts as RESTful JSON endpoints - update the prompt from PlayFetch whenever you need to. When it’s live you can monitor request performance, check how much you’re spending and more.

Deployment analytics

Collaborative Prompt Creation

With version control, commenting, labels, response histories and more, you can develop prompts alongside the rest of your team without all the back and forth.

Prompt version history
Timeline of prompts

Testing & iteration

Know when you’re making improvements. Populate test data for your prompts, test different models, and feed production requests back into your tests.

Deployment & monitoring

Expose your prompts as RESTful JSON endpoints - update the prompt from PlayFetch whenever you need to. When it’s live you can monitor request performance, check how much you’re spending and more.

A simple RESTful JSON API

The PlayFetch API is consistent, no matter which model provider you use. Every prompt or chain is exposed as its own endpoint so you never leak your API key or full access to a foundation model, and all calls are nicely self contained.

Simple Endpoint

Calling endpoints is easy - just pass inputs in the POST body. Here’s how you call a single prompt that returns immediately.

curl -X POST https://playfetch.ai/33928374234/
recipeWithIngredientsAndCalories \
-H "x-api-key: sk-x" \
-d $'{ "ingredients": "chicken, pesto",
"calories": "50" }'
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

Continuation Example

If you want to implement chat we’ll handle the state for you, just pass us a continuation key to reference the conversation like this

curl -X POST https://playfetch.ai/99388472363/
customerSupportChat \
-H "x-api-key: sk-x" \
-H "x-continuation-key: conversationID" \
-d $'{ "message": "hey my computer is on fire" }'
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

Functions

The same works for functions in OpenAI except this time we’ll pass you a continuation key which you should include when you’re done calling your function (just think of them like callbacks in your code).

Chat Example

Here’s a simple chat UI example you can link to a PlayFetch endpoint fast. View article.