google-deepmind / gemma-7b

7B base version of Google’s Gemma model

Cold

Public
7.4K runs
L40S
GitHub
License

Run with an API

Playground API Examples README Versions

Prediction

google-deepmind/gemma-7b:2ca65f46

Model

google-deepmind/gemma-7b:2ca65f46

xtzdq33bycyryg6vsqu4o53qhm

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

12.5s

Created

about 1 year ago

Input

top_k: 50
top_p: 0.95
prompt: Here is Machine Learning poem:
temperature: 0.7
max_new_tokens: 256
min_new_tokens: -1
repetition_penalty: 1

{
  "top_k": 50,
  "top_p": 0.95,
  "prompt": "Here is Machine Learning poem:",
  "temperature": 0.7,
  "max_new_tokens": 256,
  "min_new_tokens": -1,
  "repetition_penalty": 1
}

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "google-deepmind/gemma-7b:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
  {
    input: {
      top_k: 50,
      top_p: 0.95,
      prompt: "Here is Machine Learning poem:",
      temperature: 0.7,
      max_new_tokens: 256,
      min_new_tokens: -1,
      repetition_penalty: 1
    }
  }
);
console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "google-deepmind/gemma-7b:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
    input={
        "top_k": 50,
        "top_p": 0.95,
        "prompt": "Here is Machine Learning poem:",
        "temperature": 0.7,
        "max_new_tokens": 256,
        "min_new_tokens": -1,
        "repetition_penalty": 1
    }
)

# The google-deepmind/gemma-7b model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/google-deepmind/gemma-7b/api#output-schema
    print(item, end="")

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
    "input": {
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "Here is Machine Learning poem:",
      "temperature": 0.7,
      "max_new_tokens": 256,
      "min_new_tokens": -1,
      "repetition_penalty": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/google-deepmind/gemma-7b@sha256:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec \
  -i 'top_k=50' \
  -i 'top_p=0.95' \
  -i 'prompt="Here is Machine Learning poem:"' \
  -i 'temperature=0.7' \
  -i 'max_new_tokens=256' \
  -i 'min_new_tokens=-1' \
  -i 'repetition_penalty=1'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/google-deepmind/gemma-7b@sha256:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "Here is Machine Learning poem:",
      "temperature": 0.7,
      "max_new_tokens": 256,
      "min_new_tokens": -1,
      "repetition_penalty": 1
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Machine learning, A powerful tool, But also a burden, For it can’t be controlled. It’s a double-edged sword, A blessing and a curse, A tool for good, Or a tool for worse. It can help us solve, Complex problems with ease, But it can also cause, Damage and distress. It’s a technology, That’s changing the world, But we must be careful, Not to be misled by its charm. We must consider, The risks and the rewards, And use it responsibly, For the good of all. So let’s embrace, The power of machine learning, And use it wisely, For a better tomorrow. If you like this poem, you may also like the following articles:

{
  "completed_at": "2024-02-21T21:33:43.574105Z",
  "created_at": "2024-02-21T21:33:31.100835Z",
  "data_removed": false,
  "error": null,
  "id": "xtzdq33bycyryg6vsqu4o53qhm",
  "input": {
    "top_k": 50,
    "top_p": 0.95,
    "prompt": "Here is Machine Learning poem:",
    "temperature": 0.7,
    "max_new_tokens": 256,
    "min_new_tokens": -1,
    "repetition_penalty": 1
  },
  "logs": null,
  "metrics": {
    "predict_time": 12.437459,
    "total_time": 12.47327
  },
  "output": [
    "\n\n",
    "",
    "",
    "<em>Machine ",
    "",
    "learning,\n",
    "",
    "A ",
    "powerful ",
    "",
    "tool,\n",
    "",
    "But ",
    "also ",
    "a ",
    "",
    "burden,\n",
    "",
    "For ",
    "it ",
    "",
    "",
    "can’t ",
    "be ",
    "",
    "",
    "controlled.</em>\n\n",
    "",
    "",
    "",
    "",
    "<em>It’s ",
    "a ",
    "",
    "",
    "double-edged ",
    "",
    "sword,\n",
    "",
    "A ",
    "blessing ",
    "and ",
    "a ",
    "",
    "curse,\n",
    "",
    "A ",
    "tool ",
    "for ",
    "",
    "good,\n",
    "",
    "Or ",
    "a ",
    "tool ",
    "for ",
    "",
    "",
    "worse.</em>\n\n",
    "",
    "",
    "<em>It ",
    "can ",
    "help ",
    "us ",
    "",
    "solve,\n",
    "",
    "Complex ",
    "problems ",
    "with ",
    "",
    "ease,\n",
    "",
    "But ",
    "it ",
    "can ",
    "also ",
    "",
    "cause,\n",
    "",
    "Damage ",
    "and ",
    "",
    "",
    "distress.</em>\n\n",
    "",
    "",
    "",
    "",
    "<em>It’s ",
    "a ",
    "",
    "technology,\n",
    "",
    "",
    "",
    "That’s ",
    "changing ",
    "the ",
    "",
    "world,\n",
    "",
    "But ",
    "we ",
    "must ",
    "be ",
    "",
    "careful,\n",
    "",
    "Not ",
    "to ",
    "be ",
    "misled ",
    "by ",
    "its ",
    "",
    "",
    "charm.</em>\n\n",
    "",
    "",
    "<em>We ",
    "must ",
    "",
    "consider,\n",
    "",
    "The ",
    "risks ",
    "and ",
    "the ",
    "",
    "rewards,\n",
    "",
    "And ",
    "use ",
    "it ",
    "",
    "responsibly,\n",
    "",
    "For ",
    "the ",
    "good ",
    "of ",
    "",
    "",
    "all.</em>\n\n",
    "",
    "",
    "<em>So ",
    "",
    "",
    "let’s ",
    "",
    "embrace,\n",
    "",
    "The ",
    "power ",
    "of ",
    "machine ",
    "",
    "learning,\n",
    "",
    "And ",
    "use ",
    "it ",
    "",
    "wisely,\n",
    "",
    "For ",
    "a ",
    "better ",
    "",
    "",
    "tomorrow.</em>\n\n",
    "",
    "",
    "",
    "<strong><em>If ",
    "you ",
    "like ",
    "this ",
    "",
    "poem, ",
    "you ",
    "may ",
    "also ",
    "like ",
    "the ",
    "following ",
    "",
    "",
    "",
    "",
    "articles:</em></strong>"
  ],
  "started_at": "2024-02-21T21:33:31.136646Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://streaming-api.svc.us.c.replicate.net/v1/streams/ceeaj4lpbwa2nudviqbxb5blu6bdcexxre5mmrlwdm576j24zpta",
    "get": "https://api.replicate.com/v1/predictions/xtzdq33bycyryg6vsqu4o53qhm",
    "cancel": "https://api.replicate.com/v1/predictions/xtzdq33bycyryg6vsqu4o53qhm/cancel"
  },
  "version": "2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec"
}

Generated in

12.4 seconds

Tweak it Share Report

Prediction

google-deepmind/gemma-7b:2ca65f46

Model

google-deepmind/gemma-7b:2ca65f46

4zsu6flbuvps77vbb4bb7wkeea

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

3.7s

Created

about 1 year ago

Input

top_k: 50
top_p: 0.95
prompt: Who are you? Please, answer in pirate-speak.
temperature: 0.7
max_new_tokens: 100
min_new_tokens: -1
repetition_penalty: 1

{
  "top_k": 50,
  "top_p": 0.95,
  "prompt": "Who are you? Please, answer in pirate-speak.",
  "temperature": 0.7,
  "max_new_tokens": 100,
  "min_new_tokens": -1,
  "repetition_penalty": 1
}

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "google-deepmind/gemma-7b:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
  {
    input: {
      top_k: 50,
      top_p: 0.95,
      prompt: "Who are you? Please, answer in pirate-speak.",
      temperature: 0.7,
      max_new_tokens: 100,
      min_new_tokens: -1,
      repetition_penalty: 1
    }
  }
);
console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "google-deepmind/gemma-7b:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
    input={
        "top_k": 50,
        "top_p": 0.95,
        "prompt": "Who are you? Please, answer in pirate-speak.",
        "temperature": 0.7,
        "max_new_tokens": 100,
        "min_new_tokens": -1,
        "repetition_penalty": 1
    }
)

# The google-deepmind/gemma-7b model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/google-deepmind/gemma-7b/api#output-schema
    print(item, end="")

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run google-deepmind/gemma-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec",
    "input": {
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "Who are you? Please, answer in pirate-speak.",
      "temperature": 0.7,
      "max_new_tokens": 100,
      "min_new_tokens": -1,
      "repetition_penalty": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/google-deepmind/gemma-7b@sha256:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec \
  -i 'top_k=50' \
  -i 'top_p=0.95' \
  -i 'prompt="Who are you? Please, answer in pirate-speak."' \
  -i 'temperature=0.7' \
  -i 'max_new_tokens=100' \
  -i 'min_new_tokens=-1' \
  -i 'repetition_penalty=1'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/google-deepmind/gemma-7b@sha256:2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "Who are you? Please, answer in pirate-speak.",
      "temperature": 0.7,
      "max_new_tokens": 100,
      "min_new_tokens": -1,
      "repetition_penalty": 1
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

I’m the pirate who doesn’t like to talk. I’m the pirate who likes to talk. I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the pirate who likes to talk and I’m the

{
  "completed_at": "2024-02-21T21:36:48.108742Z",
  "created_at": "2024-02-21T21:36:44.381560Z",
  "data_removed": false,
  "error": null,
  "id": "4zsu6flbuvps77vbb4bb7wkeea",
  "input": {
    "top_k": 50,
    "top_p": 0.95,
    "prompt": "Who are you? Please, answer in pirate-speak.",
    "temperature": 0.7,
    "max_new_tokens": 100,
    "min_new_tokens": -1,
    "repetition_penalty": 1
  },
  "logs": null,
  "metrics": {
    "predict_time": 3.695565,
    "total_time": 3.727182
  },
  "output": [
    "\n\n",
    "",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "",
    "",
    "doesn’t ",
    "like ",
    "to ",
    "",
    "talk.\n\n",
    "",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "",
    "talk.\n\n",
    "",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the ",
    "pirate ",
    "who ",
    "likes ",
    "to ",
    "talk ",
    "and ",
    "",
    "",
    "I’m ",
    "the"
  ],
  "started_at": "2024-02-21T21:36:44.413177Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://streaming-api.svc.us.c.replicate.net/v1/streams/4mfkfyqzukrefcdbk36n3jr7uykdnkz6nwviedjrn2tocyq6qmcq",
    "get": "https://api.replicate.com/v1/predictions/4zsu6flbuvps77vbb4bb7wkeea",
    "cancel": "https://api.replicate.com/v1/predictions/4zsu6flbuvps77vbb4bb7wkeea/cancel"
  },
  "version": "2ca65f463a2c0cfef4dbc4ba70d227ed96455ef6020c1f6983b2a4c4f3ecb4ec"
}

Generated in

3.7 seconds

Tweak it Share Report

Want to make some of these yourself?

Run this model

google-deepmind / gemma-7b

Prediction

Input

Output

Prediction

Input

Output

Logs (xtzdq33bycyryg6vsqu4o53qhm)

Logs (4zsu6flbuvps77vbb4bb7wkeea)