github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-11-12 03:46:14 -06:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

docs: merging to remove retired model recommendations

Browse Source
This commit is contained in:
anra8571
2026-11-06 15:21:50 -08:00
parent d81332c7f8 c053d1e985
commit 04494de5b2
410 changed files with 44819 additions and 3622 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/api/cloud-usage.mdx
+151 -2
View File
@@ -92,9 +92,158 @@ All timestamps are in UTC. Each `from` is inclusive and each `until` is exclusiv
`partial` indicates whether the hour or day is still in progress, rather than whether all usage has been recorded.
## Model breakdown
Add `group_by=model` to see usage by model:
```shell
curl 'https://ollama.com/api/usage?range=7d&group_by=model' \
-H "Authorization: Bearer $OLLAMA_API_KEY"
```
The response includes `groups` in both `totals` and every bucket.
```json Response excerpt
{
"totals": {
"groups": [
{
"model": "glm-5.3-flash",
"request_count": 12,
"usage_usd": 0.0132
}
]
},
"buckets": [
{
"from": "2026-09-30T00:00:00Z",
"until": "2026-10-01T00:00:00Z",
"groups": [
{
"model": "glm-5.3-flash",
"request_count": 6,
"usage_usd": 0.0066
}
]
}
]
}
```
<Accordion title="Example with multiple models and buckets">
```json Response (6 empty buckets omitted)
{
"range": "7d",
"scope": "self",
"granularity": "day",
"group_by": "model",
"from": "2026-09-24T00:00:00Z",
"until": "2026-10-01T02:30:00Z",
"totals": {
"request_count": 16,
"usage_usd": 0.0268,
"input_tokens": 116000,
"cached_input_tokens": 60000,
"output_tokens": 15000,
"groups": [
{
"model": "glm-5.3-flash",
"request_count": 12,
"usage_usd": 0.0132,
"input_tokens": 96000,
"cached_input_tokens": 60000,
"output_tokens": 12000
},
{
"model": "glm-5.3",
"request_count": 2,
"usage_usd": 0.01,
"input_tokens": 4000,
"cached_input_tokens": 0,
"output_tokens": 1000
},
{
"model": "gpt-oss:120b",
"request_count": 2,
"usage_usd": 0.0036,
"input_tokens": 16000,
"cached_input_tokens": 0,
"output_tokens": 2000
}
]
},
"buckets": [
{
"from": "2026-09-30T00:00:00Z",
"until": "2026-10-01T00:00:00Z",
"request_count": 8,
"usage_usd": 0.0166,
"input_tokens": 52000,
"cached_input_tokens": 30000,
"output_tokens": 7000,
"groups": [
{
"model": "glm-5.3-flash",
"request_count": 6,
"usage_usd": 0.0066,
"input_tokens": 48000,
"cached_input_tokens": 30000,
"output_tokens": 6000
},
{
"model": "glm-5.3",
"request_count": 2,
"usage_usd": 0.01,
"input_tokens": 4000,
"cached_input_tokens": 0,
"output_tokens": 1000
}
]
},
{
"from": "2026-10-01T00:00:00Z",
"until": "2026-10-01T02:30:00Z",
"partial": true,
"request_count": 8,
"usage_usd": 0.0102,
"input_tokens": 64000,
"cached_input_tokens": 30000,
"output_tokens": 8000,
"groups": [
{
"model": "glm-5.3-flash",
"request_count": 6,
"usage_usd": 0.0066,
"input_tokens": 48000,
"cached_input_tokens": 30000,
"output_tokens": 6000
},
{
"model": "gpt-oss:120b",
"request_count": 2,
"usage_usd": 0.0036,
"input_tokens": 16000,
"cached_input_tokens": 0,
"output_tokens": 2000
}
]
}
]
}
```
</Accordion>
Currently, up to 10 groups appear under `groups`, ranked by total `usage_usd` across the range.
Every bucket uses the same selection and order. Groups with no requests in a bucket are omitted; an empty bucket has `groups: []`.
Additional groups are combined under `remainder`, omitted when empty. Totals always include all usage.
<Accordion title="Legacy plans">
For legacy plans, only request counts are available. If a bucket or total includes legacy requests, cost and token counts are omitted.
For usage that includes requests from legacy plans, only request counts are available. If the range includes legacy requests, groups are ranked by request count instead of cost.
</Accordion>
@@ -115,6 +264,6 @@ curl 'https://ollama.com/api/usage?range=7d&scope=team' \
## Coming soon
We plan to add custom time ranges, usage breakdowns by model and API key, and breakdowns by team member for team admins.
We plan to add custom time ranges, usage breakdowns by API key, and breakdowns by team member for team admins.
## Reference
docs/api/introduction.mdx
+1 -1
View File
@@ -22,7 +22,7 @@ The Anthropic client appends `/v1/messages` to its base URL.
Cloud requests need an API key. Local requests do not. To use cloud models through your local server, [sign in to Ollama](/api/authentication#signing-in).
Creating or deleting models requires a local Ollama server. [System One](/capabilities/decision) also requires a local server and is not yet available in Ollama's cloud. For supported API features, see [OpenAI compatibility](/api/openai-compatibility) and [Anthropic compatibility](/api/anthropic-compatibility).
Creating or deleting models requires a local Ollama server. For supported API features, see [OpenAI compatibility](/api/openai-compatibility) and [Anthropic compatibility](/api/anthropic-compatibility).
## Ollama API example
docs/api/openai-compatibility.mdx
+64
View File
@@ -315,6 +315,70 @@ curl -X POST http://localhost:11434/v1/chat/completions \
- [x] `dimensions`
- [ ] `user`
### `/v1/audio/speech`
> Note: Added in Ollama v0.40.8
Generates audio with [speech models](/capabilities/speech). The response streams while audio is generated.
#### Supported request fields
- [x] `model`
- [x] `input` (up to 4,096 characters)
- [x] `voice` (one of the voices `ollama show` lists, in any case and with or without accents, such as `Freya`; omit it for the model's `default` voice)
- [x] `response_format`
- [x] `wav` (default)
- [x] `pcm`
- [ ] `mp3`, `opus`, `aac`, `flac`
- [x] `speed` (`1` only)
- [x] `stream_format` (`audio` only)
- [ ] `instructions`
#### Notes
- Audio is 16-bit, mono, 24 kHz. `pcm` returns raw little-endian samples with no header.
- The default `response_format` is `wav`, not `mp3`.
- Unsupported values return a `400` error instead of being ignored.
### `/v1/audio/transcriptions`
Transcribes speech with models that accept audio input.
#### Supported request fields
- [x] `file` (WAV at up to 384 kHz; up to 512 MB and 2 hours)
- [x] `model`
- [x] `language` (ISO 639-1 code such as `fr`, or an English language name; omit it to let the model detect the language)
- [x] `prompt` (used by models without speech-specific instructions)
- [x] `response_format`
- [x] `json` (default)
- [x] `text`
- [x] `verbose_json` (one segment per transcribed piece of audio; `language` is the output language for translations; for transcriptions it is set only when given in the request)
- [ ] `srt`, `vtt`
- [x] `temperature` (default `0`)
- [x] `stream` (server-sent `transcript.text.delta` and `transcript.text.done` events)
- [ ] `timestamp_granularities`
#### Notes
- Audio longer than the model takes at once (30 seconds for current models) is split at pauses and transcribed piece by piece. Streaming sends each piece as it finishes.
- Each model gets the transcription instructions its publisher documents.
### `/v1/audio/translations`
> Note: Added in Ollama v0.40.12
Translates speech into text in another language.
#### Supported request fields
- [x] `file`
- [x] `model`
- [x] `to_language` (Ollama extension; ISO 639-1 code of the output language, default `en`)
- [x] `language` (Ollama extension; the spoken language, if known)
- [x] `prompt` (used by models without speech-specific instructions)
- [x] `response_format`, `temperature` and `stream`, as for transcriptions
### Responses API
`POST /v1/responses`
docs/capabilities/decision.mdx
+39 -4
View File
@@ -3,10 +3,6 @@ title: Decision
description: Classify text, estimate yes/no probabilities, and score ordered criteria.
---
<Note>
Currently available locally in Ollama.
</Note>
Use System One to classify text, answer yes/no questions, or score text against a rubric.
See the [full list of decision models](https://ollama.com/search?c=decision).
@@ -161,6 +157,45 @@ Example response:
}
```
## Use a cloud model
[Clef](https://ollama.com/library/clef) also runs in [Ollama's cloud](/cloud) with a paid plan or purchased [usage credits](https://ollama.com/pricing). Add `:cloud` to the model name to send the request through a local Ollama server, v0.40.5 or later, signed in to ollama.com:
```shell cloud.sh
curl http://localhost:11434/v1/systemone \
-H 'Content-Type: application/json' \
-d '{
"model": "clef:cloud",
"state": "Our checkout has returned 500 errors since 9am.",
"questions": {
"urgent": {
"type": "noul",
"instructions": "Does this ticket need attention within the hour?"
}
}
}'
```
Or call ollama.com directly with an [API key](https://ollama.com/settings/keys):
```shell cloud-direct.sh
curl https://ollama.com/v1/systemone \
-H "Authorization: Bearer $OLLAMA_API_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "clef",
"state": "Our checkout has returned 500 errors since 9am.",
"questions": {
"urgent": {
"type": "noul",
"instructions": "Does this ticket need attention within the hour?"
}
}
}'
```
Cloud requests accept up to 64 questions, named with up to 100 ASCII letters, digits, `_`, `.`, or `-`, and up to 4 images of at most 4 MiB and 16 megapixels each, 8 MiB in total. Input is limited to the model's 65,536-token context window; longer input is rejected. Usage is billed per input token at the [listed price](https://ollama.com/pricing).
## Question types
| Type | Criteria | Result |
docs/capabilities/speech.mdx
+128
View File
@@ -0,0 +1,128 @@
---
title: Speech
description: Turn text into spoken audio with text-to-speech models.
---
Speech models read text aloud. Ollama streams the audio as it is generated, so playback can start before the whole text has been spoken.
## Setup
Start [Ollama](/quickstart) v0.40.9 or later, then download [Phonon](https://ollama.com/library/phonon):
```shell setup.sh
ollama pull phonon
```
Phonon uses Gradium's 31M English checkpoint and comes with Gradium's English voices, such as `Zoey`, `Easton`, and `Freya`. Without a voice, it speaks with the checkpoint's own voice, `default`. `ollama show phonon` lists every voice. Voice names work in any case and with or without accents.
## Speak from the terminal
`ollama run` plays the audio through your computer's speakers:
```shell examples.sh
# Speak some text
ollama run phonon "Hello from Ollama."
# Choose a voice
ollama run phonon --voice Zoey "Hi, this is Zoey."
# Speak text from another program
echo "Text from another program works too." | ollama run phonon
# Speak a chat model's answer
ollama run gemma4 --hidethinking "Tell me a fun fact about owls in two sentences." | ollama run phonon
```
Piped text is spoken as it arrives, so a chat model's answer starts playing after its first sentence or two.
Run `ollama run phonon` without any text to type lines at a `>>>` prompt. Each line is spoken when you press Enter. Press Ctrl+C to stop speaking, and type `/bye` or press Ctrl+D to exit.
Audio plays with your system's built-in player: `afplay` on macOS, the first of `paplay`, `pw-play`, `aplay`, or `ffplay` that is installed on Linux, and PowerShell on Windows.
### Save audio to a file
When the output is redirected, `ollama run` writes a WAV file instead of playing it. This also works on servers without speakers:
```shell
ollama run phonon "Save this to a file." > speech.wav
```
To hear a remote server, run `ollama` on your own computer and point it at the server. The audio plays where the command runs:
```shell
OLLAMA_HOST=my-server:11434 ollama run phonon "Hello from the server."
```
## Generate speech with the API
Use the OpenAI-compatible `/v1/audio/speech` endpoint:
<Tabs>
<Tab title="cURL">
```shell
curl http://localhost:11434/v1/audio/speech \
-H "Content-Type: application/json" \
-d '{
"model": "phonon",
"input": "Hello from Ollama.",
"voice": "default"
}' --output speech.wav
```
</Tab>
<Tab title="Python">
```python
from openai import OpenAI
client = OpenAI(base_url="http://localhost:11434/v1", api_key="ollama")
with client.audio.speech.with_streaming_response.create(
model="phonon",
voice="default",
input="Hello from Ollama.",
response_format="wav",
) as response:
response.stream_to_file("speech.wav")
```
</Tab>
<Tab title="JavaScript">
```javascript
import fs from 'node:fs'
import OpenAI from 'openai'
const openai = new OpenAI({ baseURL: 'http://localhost:11434/v1', apiKey: 'ollama' })
const response = await openai.audio.speech.create({
model: 'phonon',
voice: 'default',
input: 'Hello from Ollama.',
response_format: 'wav',
})
fs.writeFileSync('speech.wav', Buffer.from(await response.arrayBuffer()))
```
</Tab>
</Tabs>
The response streams while audio is generated, so you can start playing it right away:
```shell
curl -sN http://localhost:11434/v1/audio/speech \
-H "Content-Type: application/json" \
-d '{"model": "phonon", "input": "Playback starts while the rest is still being generated."}' \
| ffplay -nodisp -autoexit -loglevel error -
```
### Audio formats
| `response_format` | Output |
| --- | --- |
| `wav` (default) | A WAV file: 16-bit, mono, 24 kHz. The header is written before the length is known, so its size fields hold the maximum value. |
| `pcm` | Raw 16-bit little-endian samples, mono, 24 kHz, with no header. |
### Limits
- `input` can be up to 4,096 characters per request. `ollama run` splits longer text at paragraph and sentence breaks for you.
- Prices, numbers, symbols, email addresses, and URLs are spelled out before speaking, so `$4.59` is read as "4 dollars and 59 cents". Times and numeric dates are read as written, so write dates out (`October 23`, not `10/23`) for a natural reading. Emoji, bullet characters, and markdown markers such as `*` and `#` are skipped.
- Options the model doesn't support, such as other formats, speeds, or `instructions`, return a `400` error instead of being ignored. An unknown voice returns a `400` error that lists the available voices.
- Speech models only generate speech: chat, completion, and embedding requests to them return a `400` error.
`ollama show phonon` and `/api/show` list the model's capabilities as `speech`, and its voices (`voices` in `/api/show`). See [OpenAI compatibility](/api/openai-compatibility) for every supported field.
docs/cli.mdx
+12
View File
@@ -96,6 +96,18 @@ ollama run nimble --ask "How urgent is this?" \
The state is piped in or given after the model name, as text or JSON. Each `--ask` question is answered yes or no, unless it's followed by labels to choose from (`--choice`) or levels to score it on (`--level`). See [Decision](/capabilities/decision#ask-from-the-terminal).
### Speak with a speech model
```
ollama run phonon "Hello from Ollama."
ollama run phonon --voice Zoey "Hi, this is Zoey."
ollama run phonon "Save this to a file." > speech.wav
```
The audio plays through your speakers, or is written as a WAV file when the output is redirected. Run it without text to speak each line you type. See [Speech](/capabilities/speech#speak-from-the-terminal).
### Download a model
```
docs/cloud.mdx
+3 -1
View File
@@ -3,7 +3,9 @@ title: Cloud
sidebarTitle: Overview
---
Run larger models in Ollama's cloud without requiring a powerful GPU. Cloud models support the same capabilities as local models, but inference runs on Ollama's cloud service.
Run larger models in Ollama's cloud without requiring a powerful GPU. Inference runs on Ollama's cloud service.
[System One](/capabilities/decision#use-a-cloud-model) is also available in the cloud for classification, yes/no questions, and scoring. Supported capabilities vary by model. Cloud models currently do not support [structured outputs](/capabilities/structured-outputs).
## Choose how to connect
docs/cloud/api.mdx
+2
View File
@@ -39,4 +39,6 @@ curl https://ollama.com/api/chat \
Read the answer from `message.content`.
For classification, yes/no questions, and scoring, call System One at `https://ollama.com/v1/systemone` with the same API key. See the [System One cloud examples](/capabilities/decision#use-a-cloud-model).
You can also use [OpenAI](/api/openai-compatibility) or [Anthropic](/api/anthropic-compatibility) clients. Each supports a subset of the original API. For endpoint schemas and response fields, see the [API Reference](/api/introduction).
docs/cloud/model-lifecycle.mdx
+3 -3
View File
@@ -46,8 +46,8 @@ No cloud model retirements are currently scheduled.
| `glm-4.7` | `glm-5.2` |
| `glm-5` | `glm-5.2` |
| `minimax-m2.1` | `minimax-m3` |
| `qwen3-coder-next` | `qwen3.5:397b` |
| `qwen3-coder:480b` | `qwen3.5:397b` |
| `qwen3-coder-next` | |
| `qwen3-coder:480b` | |
</Accordion>
<Accordion title="June 30, 2026">
@@ -62,7 +62,7 @@ No cloud model retirements are currently scheduled.
| `kimi-k2-thinking` | `kimi-k2.6` |
| `kimi-k2:1t` | `kimi-k2.6` |
| `minimax-m2` | `minimax-m3` |
| `glm-4.6` | `glm-5.1` |
| `glm-4.6` | |
| `qwen3-next:80b` | `qwen3.5` |
| `qwen3-vl:235b` | `qwen3.5` |
| `qwen3-vl:235b-instruct` | `qwen3.5` |
docs/cloud/models.mdx
+4
View File
@@ -36,3 +36,7 @@ curl https://ollama.com/api/tags
Use the name returned by this list, such as `gemma4:31b`. In the Ollama app, CLI, official libraries, or local API, use the `:cloud` name, such as `gemma4:cloud`.
To call a model without installing Ollama, first [create an API key](/cloud/api).
## System One
Use [Clef](https://ollama.com/library/clef) for classification, yes/no questions, and scoring in the cloud. Through a signed-in local Ollama server, use `clef:cloud` with Ollama v0.40.5 or later. Direct API requests use `clef`. Cloud access requires a paid plan or purchased usage credits. See the [System One cloud examples](/capabilities/decision#use-a-cloud-model).
docs/docs.json
+1
View File
@@ -130,6 +130,7 @@
"/capabilities/decision",
"/capabilities/vision",
"/capabilities/embeddings",
"/capabilities/speech",
"/capabilities/tool-calling",
"/capabilities/web-search"
]
docs/openapi.yaml
+74 -16
View File
@@ -634,6 +634,15 @@ components:
template:
type: string
description: The template used by the model to render prompts
renderer:
type: string
description: The effective chat renderer of the model (aliases resolved), when one is set
parser:
type: string
description: The output parser configured on the model, when one is set
template_source:
type: string
description: Where the chat prompt actually comes from; one of `renderer_parser`, `renderer`, `parser` (a Go renderer runs and any packaged template is inert), `harmony`, `go_template`, `gguf_chat_template` (the runner renders the GGUF's embedded jinja), or `none`
capabilities:
type: array
items:
@@ -899,18 +908,18 @@ components:
model:
type: string
pattern: '\S'
description: Local model trained for System One, such as nimble.
description: Model trained for System One, such as nimble, or clef:cloud for Ollama's cloud.
state:
$ref: "#/components/schemas/SystemOneContent"
images:
type: array
description: Base64-encoded images shared by all questions, in request order. Requires a model with vision, such as Clef. URLs and data URLs are not supported.
description: Base64-encoded images shared by all questions, in request order. Requires a model with vision, such as Clef. Models with audio input also accept WAV clips here. URLs and data URLs are not supported.
items:
type: string
format: byte
questions:
type: object
description: Named questions about the shared state. Answers are not passed to later questions.
description: Named questions about the shared state. Answers are not passed to later questions. Cloud models accept names of up to 100 ASCII letters, digits, `_`, `.`, or `-`.
minProperties: 1
maxProperties: 64
propertyNames:
@@ -1035,9 +1044,31 @@ components:
output_tokens:
type: integer
description: Recorded output tokens.
UsageBucket:
UsageGroup:
allOf:
- $ref: "#/components/schemas/UsageMetrics"
- type: object
required: [model]
properties:
model:
type: string
description: Model name.
UsageSummary:
allOf:
- $ref: "#/components/schemas/UsageMetrics"
- type: object
properties:
groups:
type: array
description: Present when group_by is requested. Currently returns up to 10 groups selected across the entire range, ranked by usage_usd or by request_count when the range includes legacy requests. Groups use the same order in every bucket; groups without requests in a bucket are omitted.
items:
$ref: "#/components/schemas/UsageGroup"
remainder:
$ref: "#/components/schemas/UsageMetrics"
description: Combined usage from groups outside the current limit. Omitted when there are no requests from additional groups.
UsageBucket:
allOf:
- $ref: "#/components/schemas/UsageSummary"
- type: object
required: [from, until]
properties:
@@ -1068,6 +1099,10 @@ components:
type: string
enum: [hour, day]
description: "Set automatically: hour for 24h, day for 7d and 30d."
group_by:
type: string
enum: [model]
description: Selected grouping. Present when group_by is requested.
from:
type: string
format: date-time
@@ -1077,8 +1112,8 @@ components:
format: date-time
description: End of the range in UTC, exclusive.
totals:
$ref: "#/components/schemas/UsageMetrics"
description: Usage across the entire range.
$ref: "#/components/schemas/UsageSummary"
description: Usage across the entire range, including groups in the remainder.
buckets:
type: array
description: Usage in chronological order, including buckets with no recorded requests.
@@ -1155,20 +1190,25 @@ paths:
post:
summary: Answer typed questions
description: |
Answer choice, yes/no, and scoring questions with a local System One model. Requires Ollama v0.35.0 or later. See the [decision guide](/capabilities/decision) for examples.
Answer choice, yes/no, and scoring questions with a System One model, locally or in Ollama's cloud. Requires Ollama v0.35.0 or later, or v0.40.5 or later for cloud models. See the [decision guide](/capabilities/decision) for examples.
operationId: systemone
servers:
- url: http://localhost:11434
description: Ollama
- url: https://ollama.com
description: Ollama's cloud
security:
- {}
- bearerAuth: []
x-mint:
href: /api/systemone
metadata:
title: System One
sidebarTitle: /v1/systemone
content: |
<Note>
Currently available locally in Ollama.
</Note>
- Returns one JSON response; streaming, video, tools, and generation controls are not supported.
- Requests without images must fit within 64 KiB; requests with images must fit within 32 MiB, including base64 and JSON. The complete input must fit the loaded context window. Nimble and Tev also require two token positions for scoring. Input is never truncated.
- Local requests' text and schema, excluding images, must fit within 64 KiB, and the whole request within 32 MiB, including base64 and JSON. The complete input must fit the loaded context window. Nimble and Tev also require two token positions for scoring. Input is never truncated.
- Cloud models need a paid plan or purchased [usage credits](https://ollama.com/pricing). Cloud requests accept up to 64 questions, named with up to 100 ASCII letters, digits, `_`, `.`, or `-`, and up to 4 images of at most 4 MiB and 16 megapixels each, 8 MiB in total. Requests must fit within 13 MiB, and input must fit the model's 65,536-token context window.
x-codeSamples:
- lang: bash
label: Choose a label
@@ -1229,25 +1269,37 @@ paths:
input_tokens: 174
output_tokens: 1
"400":
description: Invalid request, unsupported model or runner, or a rendered prompt that exceeds the loaded context. Cloud models are rejected.
description: Invalid request, unsupported model or runner, or a rendered prompt that exceeds the loaded context or a cloud model's token limit.
content:
application/json:
schema:
$ref: "#/components/schemas/ErrorResponse"
"401":
description: A cloud model was requested without signing in, or with an invalid API key. Through a local server, the response includes a `signin_url`.
content:
application/json:
schema:
$ref: "#/components/schemas/ErrorResponse"
"402":
description: The cloud model isn't included in the Free plan. Add usage credits or upgrade. Also returned when a payment fails.
content:
application/json:
schema:
$ref: "#/components/schemas/ErrorResponse"
"404":
description: Local model not found. Download the model before making a request.
description: Model not found. Download a local model before making a request.
content:
application/json:
schema:
$ref: "#/components/schemas/ErrorResponse"
"413":
description: Request body exceeds 64 KiB without images or 32 MiB with images.
description: Local text and schema exceed 64 KiB or the request exceeds 32 MiB, or a cloud request exceeds 13 MiB or the model's context window.
content:
application/json:
schema:
$ref: "#/components/schemas/ErrorResponse"
example:
error: request body must not exceed 64 KiB without images
error: text and schema must not exceed 64 KiB
"500":
description: Model loading, rendering, or scoring failed.
content:
@@ -2134,6 +2186,12 @@ paths:
type: string
enum: [self, team]
default: self
- name: group_by
in: query
description: "How to group usage within totals and buckets. Supported values: model."
schema:
type: string
enum: [model]
responses:
"200":
description: Usage for the selected range and scope.