Gemini and other generative AI models process input and output at a granularity
called a *token*.

**For Gemini models, a token is equivalent to about 4 characters.
100 tokens is equal to about 60-80 English words.**

## About tokens

Tokens can be single characters like `z` or whole words like `cat`. Long words
are broken up into several tokens. The set of all tokens used by the model is
called the vocabulary, and the process of splitting text into tokens is called
*tokenization*.

When billing is enabled, the [cost of a call to the Gemini API](https://ai.google.dev/pricing) is
determined in part by the number of input and output tokens, so knowing how to
count tokens can be helpful.

You can try out counting tokens in our Colab.

|---|---|---|
| [![](https://ai.google.dev/static/site-assets/images/docs/notebook-site-button.png)View on ai.google.dev](https://ai.google.dev/gemini-api/docs/tokens) | [![](https://www.tensorflow.org/images/colab_logo_32px.png)Try a Colab notebook](https://colab.research.google.com/github/google-gemini/cookbook/blob/main/quickstarts/Counting_Tokens.ipynb) | [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png)View notebook on GitHub](https://colab.research.google.com/github/google-gemini/cookbook/blob/main/quickstarts/Counting_Tokens.ipynb) |

## Count tokens

All input to and output from the Gemini API is tokenized, including text, image
files, and other non-text modalities.

You can count tokens in the following ways:

- **Call [`count_tokens`](https://ai.google.dev/api/rest/v1/models/countTokens) with the input
  of the request.**   
  This returns the total number of tokens in *the
  input only*. You can make this call before sending the input to the
  model to check the size of your requests.

- **Use the `usage_metadata` attribute on the `response` object after
  calling `generate_content`.**   
  This returns the total number of
  tokens in *both the input and the output* : `total_token_count`.  
  It
  also returns the token counts of the input and output separately:
  `prompt_token_count` (input tokens) and `candidates_token_count`
  (output tokens).

  If you are using a [thinking
  model](https://ai.google.dev/gemini-api/docs/thinking), the tokens used during the thinking
  process are returned in `thoughts_token_count`. And if you are using
  [Context caching](https://ai.google.dev/gemini-api/docs/caching), the cached token
  count will be in `cached_content_token_count`.

### Count text tokens

If you call `count_tokens` with a text-only input, it returns the token count of
the text in *the input only* (`total_tokens`). You can make this call before
calling `generate_content` to check the size of your requests.

Another option is calling `generate_content` and then using the `usage_metadata`
attribute on the `response` object to get the following:

- The separate token counts of the input (`prompt_token_count`), the cached content (`cached_content_token_count`) and the output (`candidates_token_count`)
- The token count for the thinking process (`thoughts_token_count`)
- The total number of tokens in *both the input and the output*
  (`total_token_count`)

### Python

    from google import genai

    client = genai.Client()
    prompt = "The quick brown fox jumps over the lazy dog."

    total_tokens = client.models.count_tokens(
        model="gemini-3.8-flash", contents=prompt
    )
    print("total_tokens: ", total_tokens)

    response = client.models.generate_content(
        model="gemini-3.8-flash", contents=prompt
    )

    print(response.usage_metadata)

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});
    const prompt = "The quick brown fox jumps over the lazy dog.";

    async function main() {
      const countTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: prompt,
      });
      console.log(countTokensResponse.totalTokens);

      const generateResponse = await ai.models.generateContent({
        model: "gemini-3.8-flash",
        contents: prompt,
      });
      console.log(generateResponse.usageMetadata);
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)

    // Convert prompt to a slice of *genai.Content using the helper.
    contents := []*genai.Content{
      genai.NewContentFromText(prompt, genai.RoleUser),
    }
    countResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      return err
    }
    fmt.Println("total_tokens:", countResp.TotalTokens)

    response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    usageMetadata, err := json.MarshalIndent(response.UsageMetadata, "", "  ")
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(string(usageMetadata))
        ```

### Count multi-turn (chat) tokens

If you call `count_tokens` with the chat history, it returns the total token
count of the text from each role in the chat (`total_tokens`).

Another option is calling `send_message` and then using the `usage_metadata`
attribute on the `response` object to get the following:

- The separate token counts of the input (`prompt_token_count`), the cached content (`cached_content_token_count`) and the output (`candidates_token_count`)
- The token count for the thinking process (`thoughts_token_count`)
- The total number of tokens in *both the input and the output* (`total_token_count`)

To understand how big your next conversational turn will be, you need to append
it to the history when you call `count_tokens`.

### Python

    from google import genai
    from google.genai import types

    client = genai.Client()

    chat = client.chats.create(
        model="gemini-3.8-flash",
        history=[
            types.Content(
                role="user", parts=[types.Part(text="Hi my name is Bob")]
            ),
            types.Content(role="model", parts=[types.Part(text="Hi Bob!")]),
        ],
    )

    print(
        client.models.count_tokens(
            model="gemini-3.8-flash", contents=chat.get_history()
        )
    )

    response = chat.send_message(
        message="In one sentence, explain how a computer works to a young child."
    )
    print(response.usage_metadata)

    extra = types.UserContent(
        parts=[
            types.Part(
                text="What is the meaning of life?",
            )
        ]
    )
    history = [*chat.get_history(), extra]
    print(client.models.count_tokens(model="gemini-3.8-flash", contents=history))

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});

    async function main() {
      const history = [
        { role: "user", parts: [{ text: "Hi my name is Bob" }] },
        { role: "model", parts: [{ text: "Hi Bob!" }] },
      ];
      const chat = ai.chats.create({
        model: "gemini-3.8-flash",
        history: history,
      });

      const countTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: chat.getHistory(),
      });
      console.log(countTokensResponse.totalTokens);

      const chatResponse = await chat.sendMessage({
        message: "In one sentence, explain how a computer works to a young child.",
      });
      console.log(chatResponse.usageMetadata);

      const extraMessage = {
        role: "user",
        parts: [{ text: "What is the meaning of life?" }],
      };
      const combinedHistory = [...chat.getHistory(), extraMessage];
      const combinedCountTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: combinedHistory,
      });
      console.log(
        "Combined history token count:",
        combinedCountTokensResponse.totalTokens,
      );
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)

    history := []*genai.Content{
      {Role: genai.RoleUser, Parts: []*genai.Part({Text: "Hi my name is Bob"})},
      {Role: genai.RoleModel, Parts: []*genai.Part({Text: "Hi Bob!"})},
    }
    chat, err := client.Chats.Create(ctx, "gemini-3.8-flash", nil, history)
    if err != nil {
      log.Fatal(err)
    }

    firstTokenResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", chat.History(false), nil)
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(firstTokenResp.TotalTokens)

    resp, err := chat.SendMessage(ctx, genai.NewPartFromText("In one sentence, explain how a computer works to a young child."))
    if err != nil {
      log.Fatal(err)
    }
    fmt.Printf("%#v\n", resp.UsageMetadata)

    extra := genai.NewContentFromText("What is the meaning of life?", genai.RoleUser)
    hist := chat.History(false)
    hist = append(hist, extra)

    secondTokenResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", hist, nil)
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(secondTokenResp.TotalTokens)

### Count multimodal tokens

All input to the Gemini API is tokenized, including text, image files, and other
non-text modalities. Note the following high-level key points about tokenization
of multimodal input during processing by the Gemini API:

- Image inputs with both dimensions \<=384 pixels are counted as
  258 tokens. Images larger in one or both dimensions are cropped and scaled as
  needed into tiles of 768x768 pixels, each counted as 258 tokens.

- Video and audio files are converted to tokens at the following fixed rates:
  video at 263 tokens per second and audio at 32 tokens per second. This rate
  applies to static processing (default). For agentic processing, token usage
  varies. See [Video token usage by processing mode](https://ai.google.dev/gemini-api/docs/generate-content/tokens#video-token-usage).

#### Media resolutions

[Gemini 3 models](https://ai.google.dev/gemini-api/docs/models#gemini-3) introduce granular control over
multimodal vision processing with the `media_resolution` parameter. The
`media_resolution` parameter determines the
**maximum number of tokens allocated per input image or video frame.**
Higher resolutions improve the model's ability to
read fine text or identify small details, but increase token usage and latency.

For more details about the parameter and how it can impact token calculations,
see the [media resolution](https://ai.google.dev/gemini-api/docs/generate-content/media-resolution) guide.

#### Image files

If you call `count_tokens` with a text-and-image input, it returns the combined
token count of the text and the image in *the input only* (`total_tokens`). You
can make this call before calling `generate_content` to check the size of your
requests. You can also optionally call `count_tokens` on the text and the file
separately.

Another option is calling `generate_content` and then using the `usage_metadata`
attribute on the `response` object to get the following:

- The separate token counts of the input (`prompt_token_count`), the cached content (`cached_content_token_count`) and the output (`candidates_token_count`)
- The token count for the thinking process (`thoughts_token_count`)
- The total number of tokens in *both the input and the output* (`total_token_count`)

Example that uses an uploaded image from the File API:

### Python

    from google import genai

    client = genai.Client()
    prompt = "Tell me about this image"
    your_image_file = client.files.upload(file=media / "organ.jpg")

    print(
        client.models.count_tokens(
            model="gemini-3.8-flash", contents=[prompt, your_image_file]
        )
    )

    response = client.models.generate_content(
        model="gemini-3.8-flash", contents=[prompt, your_image_file]
    )
    print(response.usage_metadata)

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});
    const prompt = "Tell me about this image";

    async function main() {
      const organ = await ai.files.upload({
        file: path.join(media, "organ.jpg"),
        config: { mimeType: "image/jpeg" },
      });

      const countTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: createUserContent([
          prompt,
          createPartFromUri(organ.uri, organ.mimeType),
        ]),
      });
      console.log(countTokensResponse.totalTokens);

      const generateResponse = await ai.models.generateContent({
        model: "gemini-3.8-flash",
        contents: createUserContent([
          prompt,
          createPartFromUri(organ.uri, organ.mimeType),
        ]),
      });
      console.log(generateResponse.usageMetadata);
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)

    file, err := client.Files.UploadFromPath(
      ctx, 
      filepath.Join(getMedia(), "organ.jpg"), 
      &genai.UploadFileConfig{
        MIMEType : "image/jpeg",
      },
    )
    if err != nil {
      log.Fatal(err)
    }
    parts := []*genai.Part{
      genai.NewPartFromText("Tell me about this image"),
      genai.NewPartFromURI(file.URI, file.MIMEType),
    }
    contents := []*genai.Content{
      genai.NewContentFromParts(parts, genai.RoleUser),
    }

    tokenResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println("Multimodal image token count:", tokenResp.TotalTokens)

    response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    usageMetadata, err := json.MarshalIndent(response.UsageMetadata, "", "  ")
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(string(usageMetadata))

Example that provides the image as inline data:

### Python

    from google import genai
    import PIL.Image

    client = genai.Client()
    prompt = "Tell me about this image"
    your_image_file = PIL.Image.open(media / "organ.jpg")

    print(
        client.models.count_tokens(
            model="gemini-3.8-flash", contents=[prompt, your_image_file]
        )
    )

    response = client.models.generate_content(
        model="gemini-3.8-flash", contents=[prompt, your_image_file]
    )
    print(response.usage_metadata)

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});
    const prompt = "Tell me about this image";
    const imageBuffer = fs.readFileSync(path.join(media, "organ.jpg"));

    const imageBase64 = imageBuffer.toString("base64");

    const contents = createUserContent([
      prompt,
      createPartFromBase64(imageBase64, "image/jpeg"),
    ]);

    async function main() {
      const countTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: contents,
      });
      console.log(countTokensResponse.totalTokens);

      const generateResponse = await ai.models.generateContent({
        model: "gemini-3.8-flash",
        contents: contents,
      });
      console.log(generateResponse.usageMetadata);
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)

    imageBytes, err := os.ReadFile("organ.jpg")
    if err != nil {
        log.Fatalf("Failed to read image file: %v", err)
    }
    parts := []*genai.Part{
      genai.NewPartFromText("Tell me about this image"),
      {
            InlineData: &genai.Blob{
                  MIMEType: "image/jpeg",
                  Data:     imageBytes,
            },
      },
    }
    contents := []*genai.Content{
      genai.NewContentFromParts(parts, genai.RoleUser),
    }

    tokenResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println("Multimodal image token count:", tokenResp.TotalTokens)

    response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    usageMetadata, err := json.MarshalIndent(response.UsageMetadata, "", "  ")
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(string(usageMetadata))

#### Video or audio files

Audio and video are each converted to tokens at the following fixed rates:

- Video: 263 tokens per second
- Audio: 32 tokens per second

If you call `count_tokens` with a text-and-video/audio input, it returns the
combined token count of the text and the video/audio file in *the input only*
(`total_tokens`). You can make this call before calling `generate_content` to
check the size of your requests. You can also optionally call `count_tokens` on
the text and the file separately.

Another option is calling `generate_content` and then using the `usage_metadata`
attribute on the `response` object to get the following:

- The separate token counts of the input (`prompt_token_count`), the cached content (`cached_content_token_count`) and the output (`candidates_token_count`)
- The token count for the thinking process (`thoughts_token_count`)
- The total number of tokens in *both the input and the output*
  (`total_token_count`).

### Python

    from google import genai
    import time

    client = genai.Client()
    prompt = "Tell me about this video"
    your_file = client.files.upload(file=media / "Big_Buck_Bunny.mp4")

    while not your_file.state or your_file.state.name != "ACTIVE":
        print("Processing video...")
        print("File state:", your_file.state)
        time.sleep(5)
        your_file = client.files.get(name=your_file.name)

    print(
        client.models.count_tokens(
            model="gemini-3.8-flash", contents=[prompt, your_file]
        )
    )

    response = client.models.generate_content(
        model="gemini-3.8-flash", contents=[prompt, your_file]
    )
    print(response.usage_metadata)

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});
    const prompt = "Tell me about this video";

    async function main() {
      let videoFile = await ai.files.upload({
        file: path.join(media, "Big_Buck_Bunny.mp4"),
        config: { mimeType: "video/mp4" },
      });

      while (!videoFile.state || videoFile.state.toString() !== "ACTIVE") {
        console.log("Processing video...");
        console.log("File state: ", videoFile.state);
        await sleep(5000);
        videoFile = await ai.files.get({ name: videoFile.name });
      }

      const countTokensResponse = await ai.models.countTokens({
        model: "gemini-3.8-flash",
        contents: createUserContent([
          prompt,
          createPartFromUri(videoFile.uri, videoFile.mimeType),
        ]),
      });
      console.log(countTokensResponse.totalTokens);

      const generateResponse = await ai.models.generateContent({
        model: "gemini-3.8-flash",
        contents: createUserContent([
          prompt,
          createPartFromUri(videoFile.uri, videoFile.mimeType),
        ]),
      });
      console.log(generateResponse.usageMetadata);
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)

    file, err := client.Files.UploadFromPath(
      ctx,
      filepath.Join(getMedia(), "Big_Buck_Bunny.mp4"),
      &genai.UploadFileConfig{
        MIMEType : "video/mp4",
      },
    )
    if err != nil {
      log.Fatal(err)
    }

    for file.State == genai.FileStateUnspecified || file.State != genai.FileStateActive {
      fmt.Println("Processing video...")
      fmt.Println("File state:", file.State)
      time.Sleep(5 * time.Second)

      file, err = client.Files.Get(ctx, file.Name, nil)
      if err != nil {
        log.Fatal(err)
      }
    }

    parts := []*genai.Part{
      genai.NewPartFromText("Tell me about this video"),
      genai.NewPartFromURI(file.URI, file.MIMEType),
    }
    contents := []*genai.Content{
      genai.NewContentFromParts(parts, genai.RoleUser),
    }

    tokenResp, err := client.Models.CountTokens(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println("Multimodal video/audio token count:", tokenResp.TotalTokens)
    response, err := client.Models.GenerateContent(ctx, "gemini-3.8-flash", contents, nil)
    if err != nil {
      log.Fatal(err)
    }
    usageMetadata, err := json.MarshalIndent(response.UsageMetadata, "", "  ")
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println(string(usageMetadata))

#### Video token usage by processing mode

Token usage for video depends on the processing mode:

| **Processing mode** | **Token calculation** | **Typical usage** |
|---|---|---|
| **Static** (default) | \~100 tokens/second by default (low resolution) or \~300 tokens/second (high resolution). All frames sampled at 1 FPS. | Predictable, proportional to video duration. |
| **Agentic** | Varies by content complexity. The model loads only the transcript and/or frames and/or audio needed to answer the prompt. | Up to 88% fewer tokens for long-form content. |

With agentic processing, a 1-hour lecture that would use \~1.08M tokens in static mode might use \~108K tokens, depending on the prompt and content.

To check actual token usage for a request, use the `usage_metadata` field in the response. In `UsageMetadata`, agentic video tokens are reported across the following fields:

- **Initial prompt** (video reference + user prompt): `prompt_token_count`
- **Navigation thinking** : `thoughts_token_count`
- **Transcript, frames, and audio loaded on demand** : `tool_use_prompt_token_count`
- **Final answer** : `candidates_token_count`

### Count thought tokens

> [!NOTE]
> **Note:** The API provides **summaries** in both the [free and paid tiers](https://ai.google.dev/gemini-api/docs/pricing). **Thought signatures** increase the input token count (and your cost) when you send them back in subsequent conversational turns.

When you turn on thinking, response pricing is the sum of output tokens and thinking tokens. You can retrieve the total number of generated thinking tokens from the `thoughtsTokenCount` field (or SDK equivalent).

### Python

    # ...
    print("Thoughts tokens:", response.usage_metadata.thoughts_token_count)
    print("Output tokens:", response.usage_metadata.candidates_token_count)

### JavaScript

    // ...
    console.log(`Thoughts tokens: ${response.usageMetadata.thoughtsTokenCount}`);
    console.log(`Output tokens: ${response.usageMetadata.candidatesTokenCount}`);

### Go

    // ...
    fmt.Println("Thoughts tokens:", response.UsageMetadata.ThoughtsTokenCount)
    fmt.Println("Output tokens:", response.UsageMetadata.CandidatesTokenCount)

Thinking models generate full thoughts to improve the quality of the final response, and then output [summaries](https://ai.google.dev/gemini-api/docs/thinking#summaries) to provide insight into the thought process. So, the API bases pricing on the full thought tokens the model generates to create a summary, even though the API only outputs the summary.

You can learn more about how to configure thinking in the [Gemini thinking](https://ai.google.dev/gemini-api/docs/thinking) guide.

## Context windows

The models available through the Gemini API have context windows that are
measured in tokens. The context window defines how much input you can provide
and how much output the model can generate. You can determine the size of the
context window by calling the [`models.get` endpoint](https://ai.google.dev/api/rest/v1/models/get)
or by looking in the [models documentation](https://ai.google.dev/gemini-api/docs/models).

### Python

    from google import genai

    client = genai.Client()
    model_info = client.models.get(model="gemini-3.8-flash")
    print(f"{model_info.input_token_limit=}")
    print(f"{model_info.output_token_limit=}")

### JavaScript

    import { GoogleGenAI } from '@google/genai';

    const ai = new GoogleGenAI({});

    async function main() {
      const modelInfo = await ai.models.get({model: 'gemini-3.8-flash'});
      console.log(modelInfo.inputTokenLimit);
      console.log(modelInfo.outputTokenLimit);
    }

    await main();

### Go

    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)
    if err != nil {
      log.Fatal(err)
    }
    modelInfo, err := client.ModelInfo(ctx, "models/gemini-3.8-flash")
    if err != nil {
      log.Fatal(err)
    }
    fmt.Println("input token limit:", modelInfo.InputTokenLimit)
    fmt.Println("output token limit:", modelInfo.OutputTokenLimit)