Skip to content

cody/modelconfig: add thinking models config examples #1103

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 11 commits into from
Apr 30, 2025
Merged

cody/modelconfig: add thinking models config examples #1103

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
3b06e29
Update model configuration documentation
taras-yemets Apr 29, 2025
296aaf5
add reasoning fields
taras-yemets Apr 29, 2025
05cc4d0
Update docs/cody/enterprise/model-config-examples.mdx
taras-yemets Apr 29, 2025
5b66b19
Update docs/cody/enterprise/model-config-examples.mdx
taras-yemets Apr 29, 2025
b9c9914
fix section name
taras-yemets Apr 29, 2025
fe64e1e
Merge branch 'ty/upd-modelconfig-examples' of github.com:sourcegraph/…
taras-yemets Apr 29, 2025
a25c155
Update docs/cody/enterprise/model-config-examples.mdx
taras-yemets Apr 29, 2025
7a4ce13
docs: update reasoning effort documentation in model configuration
taras-yemets Apr 29, 2025
1d4e8e0
maybe improve spacing
taras-yemets Apr 29, 2025
d860ad7
use actual versionID in modelRefs
taras-yemets Apr 30, 2025
504c85f
fix formatting
taras-yemets Apr 30, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
238 changes: 162 additions & 76 deletions docs/cody/enterprise/model-config-examples.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,22 +133,47 @@ Below are configuration examples for setting up various LLM providers using BYOK
],
"modelOverrides": [
{
"modelRef": "anthropic::2024-10-22::claude-3.5-sonnet",
"displayName": "Claude 3.5 Sonnet",
"modelName": "claude-3-5-sonnet-latest",
"modelRef": "anthropic::2024-10-22::claude-3-7-sonnet-latest",
"displayName": "Claude 3.7 Sonnet",
"modelName": "claude-3-7-sonnet-latest",
"capabilities": ["chat"],
"category": "accuracy",
"status": "stable",
"contextWindow": {
"maxInputTokens": 45000,
"maxOutputTokens": 4000
}
"maxInputTokens": 132000,
"maxOutputTokens": 8192
},
},
{
"modelRef": "anthropic::2024-10-22::claude-3-7-sonnet-extended-thinking",
"displayName": "Claude 3.7 Sonnet Extended Thinking",
"modelName": "claude-3-7-sonnet-latest",
"capabilities": ["chat", "reasoning"],
"category": "accuracy",
"status": "stable",
"contextWindow": {
"maxInputTokens": 93000,
"maxOutputTokens": 64000
},
"reasoningEffort": "low"
},
{
"modelRef": "anthropic::2024-10-22::claude-3-5-haiku-latest",
"displayName": "Claude 3.5 Haiku",
"modelName": "claude-3-5-haiku-latest",
"capabilities": ["autocomplete", "edit", "chat"],
"category": "speed",
"status": "stable",
"contextWindow": {
"maxInputTokens": 132000,
"maxOutputTokens": 8192
},
}
],
"defaultModels": {
"chat": "anthropic::2024-10-22::claude-3.5-sonnet",
"fastChat": "anthropic::2023-06-01::claude-3-haiku",
"codeCompletion": "fireworks::v1::deepseek-coder-v2-lite-base"
"chat": "anthropic::2024-10-22::claude-3-7-sonnet-latest",
"fastChat": "anthropic::2024-10-22::claude-3-5-haiku-latest",
"codeCompletion": "anthropic::2024-10-22::claude-3-5-haiku-latest"
}
}
```
Expand All @@ -157,8 +182,9 @@ In the configuration above,

- Set up a provider override for Anthropic, routing requests for this provider directly to the specified Anthropic endpoint (bypassing Cody Gateway)
- Add three Anthropic models:
- Two models with chat capabilities (`"anthropic::2024-10-22::claude-3.5-sonnet"` and `"anthropic::2023-06-01::claude-3-haiku"`), providing options for chat users
- One model with autocomplete capability (`"fireworks::v1::deepseek-coder-v2-lite-base"`)
- `"anthropic::2024-10-22::claude-3-7-sonnet-latest"` with chat, vision, and tools capabilities
- `"anthropic::2024-10-22::claude-3-7-sonnet-extended-thinking"` with chat and reasoning capabilities (note: to enable [Claude's extended thinking](https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking) model override should include "reasoning" capability and have "reasoningEffort" defined)
- `"anthropic::2024-10-22::claude-3-5-haiku-latest"` with autocomplete, edit, chat, and tools capabilities
- Set the configured models as default models for Cody features in the `"defaultModels"` field

</Accordion>
Expand Down Expand Up @@ -239,45 +265,61 @@ In the configuration above,
}
],
"modelOverrides": [
{
"modelRef": "openai::2024-02-01::gpt-4o",
"displayName": "GPT-4o",
"modelName": "gpt-4o",
"capabilities": ["chat"],
"category": "accuracy",
"status": "stable",
"contextWindow": {
{
"modelRef": "openai::unknown::gpt-4o",
"displayName": "GPT-4o",
"modelName": "gpt-4o",
"capabilities": ["chat"],
"category": "accuracy",
"status": "stable",
"contextWindow": {
"maxInputTokens": 45000,
"maxOutputTokens": 4000
}
},
{
"modelRef": "openai::unknown::gpt-4.1-nano",
"displayName": "GPT-4.1-nano",
"modelName": "gpt-4.1-nano",
"capabilities": ["edit", "chat", "autocomplete"],
"category": "speed",
"status": "stable",
"tier": "free",
"contextWindow": {
"maxInputTokens": 77000,
"maxOutputTokens": 16000
}
},
{
"modelRef": "openai::unknown::o3",
"displayName": "o3",
"modelName": "o3",
"capabilities": ["chat", "reasoning"],
"category": "accuracy",
"status": "stable",
"tier": "pro",
"contextWindow": {
"maxInputTokens": 68000,
"maxOutputTokens": 100000
},
"reasoningEffort": "medium"
}
},
{
"modelRef": "openai::unknown::gpt-3.5-turbo-instruct",
"displayName": "GPT-3.5 Turbo Instruct",
"modelName": "gpt-3.5-turbo-instruct",
"capabilities": ["autocomplete"],
"category": "speed",
"status": "stable",
"contextWindow": {
"maxInputTokens": 7000,
"maxOutputTokens": 4000
}
],
"defaultModels": {
"chat": "openai::unknown::gpt-4o",
"fastChat": "openai::unknown::gpt-4.1-nano",
"codeCompletion": "openai::unknown::gpt-4.1-nano"
}
],
"defaultModels": {
"chat": "openai::2024-02-01::gpt-4o",
"fastChat": "openai::2024-02-01::gpt-4o",
"codeCompletion": "openai::unknown::gpt-3.5-turbo-instruct"
}
}
```

In the configuration above,

- Set up a provider override for OpenAI, routing requests for this provider directly to the specified OpenAI endpoint (bypassing Cody Gateway)
- Add two OpenAI models:
- `"openai::2024-02-01::gpt-4o"` with "chat" capabilities - used for "chat" and "fastChat"
- `"openai::unknown::gpt-3.5-turbo-instruct"` with "autocomplete" capability - used for "autocomplete"
- Add three OpenAI models:
- `"openai::2024-02-01::gpt-4o"` with chat capability - used as a default model for chat
- `"openai::unknown::gpt-4.1-nano"` with chat, edit and autocomplete capabilities - used as a default model for fast chat and autocomplete
- `"openai::unknown::o3"` with chat and reasoning capabilities - o-series model that supports thinking, can be used for chat (note: to enable thinking, model override should include "reasoning" capability and have "reasoningEffort" defined).

</Accordion>

Expand Down Expand Up @@ -313,6 +355,33 @@ In the configuration above,
"maxOutputTokens": 4000
}
},
{
"modelRef": "azure-openai::unknown::gpt-4.1-nano",
"displayName": "GPT-4.1-nano",
"modelName": "gpt-4.1-nano",
"capabilities": ["edit", "chat", "autocomplete"],
"category": "speed",
"status": "stable",
"tier": "free",
"contextWindow": {
"maxInputTokens": 77000,
"maxOutputTokens": 16000
}
},
{
"modelRef": "azure-openai::unknown::o3-mini",
"displayName": "o3-mini",
"modelName": "o3-mini",
"capabilities": ["chat", "reasoning"],
"category": "accuracy",
"status": "stable",
"tier": "pro",
"contextWindow": {
"maxInputTokens": 68000,
"maxOutputTokens": 100000
},
"reasoningEffort": "medium"
},
{
"modelRef": "azure-openai::unknown::gpt-35-turbo-instruct-test",
"displayName": "GPT-3.5 Turbo Instruct",
Expand All @@ -328,8 +397,8 @@ In the configuration above,
],
"defaultModels": {
"chat": "azure-openai::unknown::gpt-4o",
"fastChat": "azure-openai::unknown::gpt-4o",
"codeCompletion": "azure-openai::unknown::gpt-35-turbo-instruct-test"
"fastChat": "azure-openai::unknown::gpt-4.1-nano",
"codeCompletion": "azure-openai::unknown::gpt-4.1-nano"
}
}
```
Expand All @@ -338,9 +407,11 @@ In the configuration above,

- Set up a provider override for Azure OpenAI, routing requests for this provider directly to the specified Azure OpenAI endpoint (bypassing Cody Gateway).
**Note:** For Azure OpenAI, ensure that the `modelName` matches the name defined in your Azure portal configuration for the model.
- Add two OpenAI models:
- `"azure-openai::unknown::gpt-4o"` with "chat" capability - used for "chat" and "fastChat"
- `"azure-openai::unknown::gpt-35-turbo-instruct-test"` with "autocomplete" capability - used for "autocomplete"
- Add four OpenAI models:
- `"azure-openai::unknown::gpt-4o"` with chat capability - used as a default model for chat
- `"azure-openai::unknown::gpt-4.1-nano"` with chat, edit and autocomplete capabilities - used as a default model for fast chat and autocomplete
- `"azure-openai::unknown::o3-mini"` with chat and reasoning capabilities - o-series model that supports thinking, can be used for chat (note: to enable thinking, model override should include "reasoning" capability and have "reasoningEffort" defined)
- `"azure-openai::unknown::gpt-35-turbo-instruct-test"` with "autocomplete" capability - included as an alternative model
- Since `"azure-openai::unknown::gpt-35-turbo-instruct-test"` is not supported on the newer OpenAI `"v1/chat/completions"` endpoint, we set `"useDeprecatedCompletionsAPI"` to `true` to route requests to the legacy `"v1/completions"` endpoint. This setting is unnecessary if you are using a model supported on the `"v1/chat/completions"` endpoint.

</Accordion>
Expand Down Expand Up @@ -499,48 +570,63 @@ In the configuration above,
],
"modelOverrides": [
{
"modelRef": "google::unknown::claude-3-5-sonnet",
"displayName": "Claude 3.5 Sonnet (via Google/Vertex)",
"modelName": "claude-3-5-sonnet@20240620",
"contextWindow": {
"maxInputTokens": 45000,
"maxOutputTokens": 4000
},
"capabilities": ["chat"],
"category": "accuracy",
"status": "stable"
"modelRef": "google::20250219::claude-3-7-sonnet",
"displayName": "Claude 3.7 Sonnet",
"modelName": "claude-3-7-sonnet@20250219",
"capabilities": ["chat", "vision", "tools"],
"category": "accuracy",
"status": "stable",
"contextWindow": {
"maxInputTokens": 132000,
"maxOutputTokens": 8192
}
},
{
"modelRef": "google::unknown::claude-3-haiku",
"displayName": "Claude 3 Haiku",
"modelName": "claude-3-haiku@20240307",
"capabilities": ["autocomplete", "chat"],
"category": "speed",
"status": "stable",
"contextWindow": {
"maxInputTokens": 7000,
"maxOutputTokens": 4000
}
"modelRef": "google::20250219::claude-3-7-sonnet-extended-thinking",
"displayName": "Claude 3.7 Sonnet Extended Thinking",
"modelName": "claude-3-7-sonnet@20250219",
"capabilities": ["chat", "reasoning"],
"category": "accuracy",
"status": "stable",
"reasoningEffort": "medium",
"contextWindow": {
"maxInputTokens": 93000,
"maxOutputTokens": 64000
}
},
],
"defaultModels": {
"chat": "google::unknown::claude-3-5-sonnet",
"fastChat": "google::unknown::claude-3-5-sonnet",
"codeCompletion": "google::unknown::claude-3-haiku"
}
{
"modelRef": "google::20250219::claude-3-5-haiku",
"displayName": "Claude 3.5 Haiku",
"modelName": "claude-3-5-haiku@20241022",
"capabilities": ["autocomplete", "edit", "chat", "tools"],
"category": "speed",
"status": "stable",
"contextWindow": {
"maxInputTokens": 132000,
"maxOutputTokens": 8192
}
}
],
"defaultModels": {
"chat": "google::20250219::claude-3.5-sonnet",
"fastChat": "google::20250219::claude-3-5-haiku",
"codeCompletion": "google::20250219::claude-3-5-haiku"
}
}
```

In the configuration above,

- Set up a provider override for Google Anthropic, routing requests for this provider directly to the specified endpoint (bypassing Cody Gateway)
- Add two Anthropic models:
- `"google::unknown::claude-3-5-sonnet"` with "chat" capabiity - used for "chat" and "fastChat"
- `"google::unknown::claude-3-haiku"` with "autocomplete" capability - used for "autocomplete"
- Add three Anthropic models:
- `"google::unknown::claude-3-7-sonnet"` with chat, vision, and tools capabilities
- `"google::unknown::claude-3-7-sonnet-extended-thinking"` with chat and reasoning capabilities (note: to enable [Claude's extended thinking](https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking) model override should include "reasoning" capability and have "reasoningEffort" defined)
- `"google::unknown::claude-3-5-haiku"` with autocomplete, edit, chat, and tools capabilities
- Set the configured models as default models for Cody features in the `"defaultModels"` field

</Accordion>

<Accordion title="Google Vertex (public)">
<Accordion title="Google Vertex (Gemini)">

```json
"modelConfiguration": {
Expand All @@ -559,7 +645,7 @@ In the configuration above,
"modelOverrides": [
{
"modelRef": "google::unknown::claude-3-5-sonnet",
"displayName": "Claude 3.5 Sonnet (via Google/Vertex)",
"displayName": "Claude 3.5 Sonnet (via Google Vertex)",
"modelName": "claude-3-5-sonnet@20240620",
"contextWindow": {
"maxInputTokens": 45000,
Expand Down
7 changes: 5 additions & 2 deletions docs/cody/enterprise/model-configuration.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,7 +215,7 @@ This field is an array of items, each with the following fields:
- `${apiVersionId}` specifies the API version, which helps detect compatibility issues between models and Sourcegraph instances. For example, `"2023-06-01"` can indicate that the model uses that version of the Anthropic API. If unsure, you may set this to `"unknown"` when defining custom models
- `displayName`: An optional, user-friendly name for the model. If not set, clients should display the `ModelID` part of the `modelRef` instead (not the `modelName`)
- `modelName`: A unique identifier the API provider uses to specify which model is being invoked. This is the identifier that the LLM provider recognizes to determine the model you are calling
- `capabilities`: A list of capabilities that the model supports. Supported values: **autocomplete** and **chat**
- `capabilities`: A list of capabilities that the model supports. Supported values: `autocomplete`, `chat`, `vision`, `reasoning`, `edit`, `tools`.
- `category`: Specifies the model's category with the following options:
- `"balanced"`: Typically the best default choice for most users. This category is suited for models like Sonnet 3.5 (as of October 2024)
- `"speed"`: Ideal for low-parameter models that may not suit general-purpose chat but are beneficial for specialized tasks, such as query rewriting
Expand All @@ -225,6 +225,9 @@ This field is an array of items, each with the following fields:
- `contextWindow`: An object that defines the **number of tokens** (units of text) that can be sent to the LLM. This setting influences response time and request cost and may vary according to the limits set by each LLM model or provider. It includes two fields:
- `maxInputTokens`: Specifies the maximum number of tokens for the contextual data in the prompt (e.g., question, relevant snippets)
- `maxOutputTokens`: Specifies the maximum number of tokens allowed in the response
- `reasoningEffort`: Specifies the effort on reasoning for reasoning models (having `reasoning` capability). Supported values: `high`, `medium`, `low`. How this value is treated depends on the specific provider.
For example, for Anthropic models supporting thinking, `low` effort means that the minimum [`thinking.budget_tokens`](https://docs.anthropic.com/en/api/messages#body-thinking) value (1024) will be used. For other `reasoningEffort` values, the `contextWindow.maxOutputTokens / 2` value will be used.
For OpenAI reasoning models, the `reasoningEffort` field value corresponds to the [`reasoning_effort`](https://platform.openai.com/docs/api-reference/chat/create#chat-create-reasoning_effort) request body value.
- `serverSideConfig`: Additional configuration for the model. It can be one of the following:

- `awsBedrockProvisionedThroughput`: Specifies provisioned throughput settings for AWS Bedrock models with the following fields:
Expand Down Expand Up @@ -326,7 +329,7 @@ In this modelOverrides config example:
- The model is configured to use the `"chat"` and `"reasoning"` capabilities
- The `reasoningEffort` can be set to 3 different options in the Model Config. These options are `high`, `medium` and `low`
- The default `reasoningEffort` is set to `low`
- When the reasoning effort is `low`, 1024 tokens is used as the thinking budget. With `medium` and `high` the thinking budget is set via `max_tokens_to_sample/2`
- For Anthropic models supporting thinking, when the reasoning effort is `low`, 1024 tokens is used as the thinking budget. With `medium` and `high` the thinking budget is set to half of the maxOutputTokens value

Refer to the [examples page](/cody/enterprise/model-config-examples) for additional examples.

Expand Down