Official

meta / llama-2-7b

Base version of Llama 2 7B, a 7 billion parameter language model

  • Public
  • 651.2K runs
  • Priced per token

Input

*string
Shift + Return to add a new line

Prompt to send to the model.

integer
(minimum: 1)

Maximum number of tokens to generate. A word is generally 2-3 tokens.

Default: 512

integer
(minimum: -1)

Minimum number of tokens to generate. To disable, set to -1. A word is generally 2-3 tokens.

number
(minimum: 0, maximum: 5)

Adjusts randomness of outputs, greater than 1 is random and 0 is deterministic, 0.75 is a good starting value.

Default: 0.7

number
(minimum: 0, maximum: 1)

When decoding text, samples from the top p percentage of most likely tokens; lower to ignore less likely tokens.

Default: 0.95

integer
(minimum: -1)

When decoding text, samples from the top k most likely tokens; lower to ignore less likely tokens.

Default: 0

string
Shift + Return to add a new line

A comma-separated list of sequences to stop generation at. For example, '<end>,<stop>' will stop generation at the first instance of 'end' or '<stop>'.

number
(minimum: 0, maximum: 5)

A parameter that controls how long the outputs are. If < 1, the model will tend to generate shorter outputs, and > 1 will tend to generate longer outputs.

Default: 1

number

A parameter that penalizes repeated tokens regardless of the number of appearances. As the value increases, the model will be less likely to repeat tokens in the output.

Default: 0

integer

Random seed. Leave blank to randomize the seed.

string
Shift + Return to add a new line

Template for formatting the prompt. Can be an arbitrary string, but must contain the substring `{prompt}`.

Default: "{prompt}"

boolean

Default: false

integer
(minimum: 1)

This parameter has been renamed to max_tokens. max_new_tokens only exists for backwards compatibility purposes. We recommend you use max_tokens instead. Both may not be specified.

integer
(minimum: -1)

This parameter has been renamed to min_tokens. min_new_tokens only exists for backwards compatibility purposes. We recommend you use min_tokens instead. Both may not be specified.

Output

, he orders a martini. everyone in the place stops and stares at him," said Jude Lawless: In 1983 he moved to New York City to pursue his performing career; there in November of that year he opened for Paul Simon on tour with Kate Pierson & The B-Sides (formerly of The B-52's). He also appeared with artists such as Miles Davis and Lou Reed who taught him how to write music. In February, 1984 they won an ABC-TV New Day competition contest and released their first single "That's All Right" on Scorpio Records which became so successful that it went Gold in Canada. His latest album "Tales Told By A Dead Man" has received top reviews and is being praised by many critics since its release in September, 2007. Since then, Pansy started singing in local bands including several funk groups where he began learning how to play drums with a backyard bamboo set given to him from his grandmother Mary Beth. As time passed on, Mike developed an interest towards punk rock which sparked him later to form a band called Broken Arrow (Bass/Vocals) originally started in South Los Angeles before playing clubs around Hollywood such as The Roxy Theater or The Whiskey Showroom where Bruce Springsteen was often seen patronizing after his shows according to rumors told by fans themselves! Meanwhile at home while rehearsing one night outside the house listening to The Police and The Who's "Wonka", Mike discovered something about himself...that what felt right seemed wrong because all these great songs sound good no matter who sings them even though sometimes certain voices were better suited than others like Billy Idol when covering David Lee Roth era Van Halen material on TV show or Dweezil Zappa on radio stations too much. This thought process made sense until some day when doing karaoke came up during which his friends said oh yeah dude we need you tonight because we have not heard anyone else sing them yet!. Right? So why ask me if I want another drink huh. He will say thank u, here ya go.. Jake Johnson who is also an actor can be seen recurring roles of Mr. Dexter as the owner for Dog & Scout was chosen specifically
Generated in

Pricing

Official model
Pricing for official models works differently from other models. Instead of being billed by time, you’re billed by input and output, making pricing more predictable.

This language model is priced by how many input tokens are sent as inputs and how many output tokens are generated.

TypePer unitPer $1
Input
$0.05 / 1M tokens
or
20M tokens / $1
Output
$0.25 / 1M tokens
or
4M tokens / $1

For example, for $10 you can run around 57,143 predictions where the input is a sentence or two (15 tokens) and the output is a few paragraphs (700 tokens).

Check out our docs for more information about how per-token pricing works on Replicate.