To generate text completions for a given prompt, you can send an HTTP request to the complete endpoint of your desired Jurassic-2 language model. This request should include the input text prompt as well as any relevant parameters for controlling the text generation. Additionally, you must include your API key in the request headers for authentication purposes.
Once you've sent your complete request, you'll receive a response that includes the tokenized prompt, as well as the generated text completion(s). This generated text is referred to as completions, and can include multiple outputs depending on the parameters you set. The response may also contain additional metadata relevant to the generation process.
Jurassic-2 Models
Jurassic-2 models come in three variations - Large, Grande and Jumbo.
Large: Designed for fast responses, this model can be fine-tuned to optimize performance for relatively simple tasks.
Grande: This model offers enhanced text generation capabilities, making it well-suited to language tasks with a greater degree of complexity.
Jumbo: As the largest and most powerful model in the Jurassic series, J2-Jumbo is an ideal choice for the most complex language processing tasks and generative text applications.
API Parameters
Each of these parameters can be used to customize the behavior of the Jurassic-2 language models, allowing developers to fine-tune the model's responses to better suit their needs:
prompt: The initial prompt to generate responses from.
numResults: The number of responses to generate for a given prompt.
maxTokens: The maximum number of tokens to generate for each response.
minTokens: The minimum number of tokens to generate for each response.
temperature: A value controlling the "creativity" of the model's responses.
topP: A value controlling the diversity of the model's responses.
stopSequences: A list of sequences that, when generated, will cause the model to stop generating tokens.
topKReturn: The number of top-scoring tokens to consider for each generation step.
frequencyPenalty: A penalty applied to tokens that are frequently generated.
presencePenalty: A penalty applied to tokens that are already present in the prompt.
countPenalty: A penalty applied to tokens based on their frequency in the generated responses.
Example API Request
fetch("https://api.ai21.com/studio/v1/j2-grande/complete", {
headers: {
"Authorization": "Bearer YOUR_API_KEY",
"Content-Type": "application/json"
},
body: JSON.stringify({
"prompt": "These are a few of my favorite",
"numResults": 1,
"maxTokens": 30,
"temperature": 0.7,
"topKReturn": 0,
"topP":1,
"countPenalty": {
"scale": 0,
"applyToNumbers": false,
"applyToPunctuations": false,
"applyToStopwords": false,
"applyToWhitespaces": false,
"applyToEmojis": false
},
"frequencyPenalty": {
"scale": 0,
"applyToNumbers": false,
"applyToPunctuations": false,
"applyToStopwords": false,
"applyToWhitespaces": false,
"applyToEmojis": false
},
"presencePenalty": {
"scale": 0,
"applyToNumbers": false,
"applyToPunctuations": false,
"applyToStopwords": false,
"applyToWhitespaces": false,
"applyToEmojis": false
},
"stopSequences":["##"]
}),
method: "POST"
});
Example Response
{
"id": "75a54d20-dda1-6db7-ff99-59b1af4fdfd3",
"prompt": {
"text": "These are a few of my favorite",
"tokens": [
{
"generatedToken": {
"token": "▁These▁are",
"logprob": -8.139219284057617,
"raw_logprob": -8.139219284057617
},
"topTokens": null,
"textRange": {
"start": 0,
"end": 9
}
},
{
"generatedToken": {
"token": "▁a▁few",
"logprob": -4.903847694396973,
"raw_logprob": -4.903847694396973
},
"topTokens": null,
"textRange": {
"start": 9,
"end": 15
}
},
{
"generatedToken": {
"token": "▁of▁my▁favorite",
"logprob": -1.543039321899414,
"raw_logprob": -1.543039321899414
},
"topTokens": null,
"textRange": {
"start": 15,
"end": 30
}
}
]
},
"completions": [
{
"data": {
"text": " things!\n\nI love entertaining, entertaining and decorating my home, entertaining clients, entertaining friends, entertaining family...you get the point! One of my favorite things to do is plan parties",
"tokens": [
{
"generatedToken": {
"token": "▁things",
"logprob": -0.012797726318240166,
"raw_logprob": -0.24019983410835266
},
"topTokens": null,
"textRange": {
"start": 0,
"end": 7
}
},
{
"generatedToken": {
"token": "!",
"logprob": -2.060641050338745,
"raw_logprob": -2.189206600189209
},
"topTokens": null,
"textRange": {
"start": 7,
"end": 8
}
},
{
"generatedToken": {
"token": "<|newline|>",
"logprob": -0.8403825759887695,
"raw_logprob": -2.1910078525543213
},
"topTokens": null,
"textRange": {
"start": 8,
"end": 9
}
},
{
"generatedToken": {
"token": "<|newline|>",
"logprob": -0.1456860452890396,
"raw_logprob": -0.9437531232833862
},
"topTokens": null,
"textRange": {
"start": 9,
"end": 10
}
},
{
"generatedToken": {
"token": "▁I▁love",
"logprob": -4.374380588531494,
"raw_logprob": -4.643549919128418
},
"topTokens": null,
"textRange": {
"start": 10,
"end": 16
}
},
{
"generatedToken": {
"token": "▁entertaining",
"logprob": -8.783716201782227,
"raw_logprob": -8.083599090576172
},
"topTokens": null,
"textRange": {
"start": 16,
"end": 29
}
},
{
"generatedToken": {
"token": ",",
"logprob": -1.0627485513687134,
"raw_logprob": -1.545505404472351
},
"topTokens": null,
"textRange": {
"start": 29,
"end": 30
}
},
{
"generatedToken": {
"token": "▁entertaining",
"logprob": -4.874294757843018,
"raw_logprob": -4.618663787841797
},
"topTokens": null,
"textRange": {
"start": 30,
"end": 43
}
},
{
"generatedToken": {
"token": "▁and",
"logprob": -3.167405605316162,
"raw_logprob": -3.570988655090332
},
"topTokens": null,
"textRange": {
"start": 43,
"end": 47
}
},
{
"generatedToken": {
"token": "▁decorating",
"logprob": -1.648970127105713,
"raw_logprob": -2.582613706588745
},
"topTokens": null,
"textRange": {
"start": 47,
"end": 58
}
},
{
"generatedToken": {
"token": "▁my▁home",
"logprob": -2.5026865005493164,
"raw_logprob": -2.6586031913757324
},
"topTokens": null,
"textRange": {
"start": 58,
"end": 66
}
},
{
"generatedToken": {
"token": ",",
"logprob": -0.7689361572265625,
"raw_logprob": -1.2174654006958008
},
"topTokens": null,
"textRange": {
"start": 66,
"end": 67
}
},
{
"generatedToken": {
"token": "▁entertaining",
"logprob": -1.2487196922302246,
"raw_logprob": -2.0927321910858154
},
"topTokens": null,
"textRange": {
"start": 67,
"end": 80
}
},
{
"generatedToken": {
"token": "▁clients",
"logprob": -8.5693359375,
"raw_logprob": -7.141331672668457
},
"topTokens": null,
"textRange": {
"start": 80,
"end": 88
}
},
{
"generatedToken": {
"token": ",",
"logprob": -0.1822780817747116,
"raw_logprob": -0.6432795524597168
},
"topTokens": null,
"textRange": {
"start": 88,
"end": 89
}
},
{
"generatedToken": {
"token": "▁entertaining",
"logprob": -0.5468096137046814,
"raw_logprob": -1.3048734664916992
},
"topTokens": null,
"textRange": {
"start": 89,
"end": 102
}
},
{
"generatedToken": {
"token": "▁friends",
"logprob": -0.9710591435432434,
"raw_logprob": -1.7686699628829956
},
"topTokens": null,
"textRange": {
"start": 102,
"end": 110
}
},
{
"generatedToken": {
"token": ",",
"logprob": -0.14025729894638062,
"raw_logprob": -0.4739122986793518
},
"topTokens": null,
"textRange": {
"start": 110,
"end": 111
}
},
{
"generatedToken": {
"token": "▁entertaining",
"logprob": -0.2804214358329773,
"raw_logprob": -0.8527922034263611
},
"topTokens": null,
"textRange": {
"start": 111,
"end": 124
}
},
{
"generatedToken": {
"token": "▁family",
"logprob": -1.0800895690917969,
"raw_logprob": -1.927623987197876
},
"topTokens": null,
"textRange": {
"start": 124,
"end": 131
}
},
{
"generatedToken": {
"token": "...",
"logprob": -2.514098644256592,
"raw_logprob": -2.373312473297119
},
"topTokens": null,
"textRange": {
"start": 131,
"end": 134
}
},
{
"generatedToken": {
"token": "you",
"logprob": -1.0778509378433228,
"raw_logprob": -1.99626624584198
},
"topTokens": null,
"textRange": {
"start": 134,
"end": 137
}
},
{
"generatedToken": {
"token": "▁get",
"logprob": -0.03242425248026848,
"raw_logprob": -0.1461254507303238
},
"topTokens": null,
"textRange": {
"start": 137,
"end": 141
}
},
{
"generatedToken": {
"token": "▁the▁point",
"logprob": -2.4119038581848145,
"raw_logprob": -2.0218327045440674
},
"topTokens": null,
"textRange": {
"start": 141,
"end": 151
}
},
{
"generatedToken": {
"token": "!",
"logprob": -1.4457058906555176,
"raw_logprob": -1.3022156953811646
},
"topTokens": null,
"textRange": {
"start": 151,
"end": 152
}
},
{
"generatedToken": {
"token": "▁One▁of▁my▁favorite",
"logprob": -5.320619583129883,
"raw_logprob": -5.056370258331299
},
"topTokens": null,
"textRange": {
"start": 152,
"end": 171
}
},
{
"generatedToken": {
"token": "▁things▁to▁do",
"logprob": -0.678179919719696,
"raw_logprob": -1.1683768033981323
},
"topTokens": null,
"textRange": {
"start": 171,
"end": 184
}
},
{
"generatedToken": {
"token": "▁is",
"logprob": -0.08321860432624817,
"raw_logprob": -0.5311100482940674
},
"topTokens": null,
"textRange": {
"start": 184,
"end": 187
}
},
{
"generatedToken": {
"token": "▁plan",
"logprob": -2.2561252117156982,
"raw_logprob": -3.04398512840271
},
"topTokens": null,
"textRange": {
"start": 187,
"end": 192
}
},
{
"generatedToken": {
"token": "▁parties",
"logprob": -2.2673985958099365,
"raw_logprob": -2.537139892578125
},
"topTokens": null,
"textRange": {
"start": 192,
"end": 200
}
}
]
},
"finishReason": {
"reason": "length",
"length": 30
}
}
]
}
Response
The response is a nested data structure containing information about the processed request, prompt, and completions. At the top level, the response has the following fields:
ID
A unique string id for the processed request. Repeated identical requests receive different IDs.
prompt
The prompt includes the raw text, the tokens with their log probabilities, and the top-K alternative tokens at each position, if requested. It has two nested fields:
text(string)tokens(list of TokenData)
completions
A list of completions, including raw text, tokens, and log probabilities. The number of completions corresponds to the requested numResults. Each completion has two nested fields:
data, which contains thetext(string) and tokens (list of TokenData) for the completion.finishReason, a nested data structure describing the reason generation was terminated for this completion.
TokenData
The TokenData object provides detailed information about each token in both the prompt and the completions. It includes the following fields:
generatedToken:
The generatedToken field consists of two nested fields:
token: The string representation of the token.logprob: The predicted log probability of the token as a float value..
topTokens
The topTokens field is a list of the top K alternative tokens for this position, sorted by probability, according to the topKReturn request parameter. If topKReturn is set to 0, this field will be null.
Each token in the list includes:
token: The string representation of the alternative token.logprob: The predicted log probability of the alternative token as a float value.
textRange
The textRange field indicates the start and end offsets of the token in the decoded text string:
start: The starting index of the token in the decoded text string.end: The ending index of the token in the decoded text string.