From 9524db3fb77da247b3c477e2555d06b2478b1f86 Mon Sep 17 00:00:00 2001 From: fpagny Date: Fri, 10 Oct 2025 16:54:31 +0200 Subject: [PATCH 1/4] feat(genapi): update Continue suggested models --- .../adding-ai-to-vscode-using-continue.mdx | 22 +++++++++---------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/pages/generative-apis/reference-content/adding-ai-to-vscode-using-continue.mdx b/pages/generative-apis/reference-content/adding-ai-to-vscode-using-continue.mdx index 49cf4d76df..23114f8d8a 100644 --- a/pages/generative-apis/reference-content/adding-ai-to-vscode-using-continue.mdx +++ b/pages/generative-apis/reference-content/adding-ai-to-vscode-using-continue.mdx @@ -56,21 +56,21 @@ To link Continue with Scaleway's Generative APIs, you can configure a settings f - If you have already configured a **Local Assistant**, click **Local Assistant**, then click the **wheel icon** to open your existing `config.yaml` - Otherwise, create a `config.yaml` file inside your `.continue` directory. - Add the following configuration to enable Scaleway's Generative API. This configuration uses three different models for each tasks: - - `devstral-small-2505` for agentic workflows through a chat interface - - `qwen2.5-coder-32b-instruct` for autocompletion when editing a file + - `qwen3-235b-a22b-instruct-2507` for agentic workflows through a chat interface (`qwen3-coder-30b-a3b-instruct` can also be used) + - `qwen3-coder-30b-a3b-instruct` for autocompletion when editing a file - `bge-multilingual-gemma2` for embedding and retrieving code context ```yaml name: Continue Config version: 0.0.1 models: - - name: Devstral - Scaleway + - name: Qwen3 235B - Scaleway provider: openai - model: devstral-small-2505 + model: qwen3-235b-a22b-instruct-2507 apiBase: https://api.scaleway.ai/v1/ apiKey: ###SCW_SECRET_KEY### defaultCompletionOptions: - maxTokens: 8000 - contextLength: 50000 + maxTokens: 4000 + contextLength: 40000 roles: - chat - apply @@ -80,12 +80,12 @@ To link Continue with Scaleway's Generative APIs, you can configure a settings f - tool_use - name: Autocomplete - Scaleway provider: openai - model: qwen2.5-coder-32b-instruct + model: qwen3-coder-30b-a3b-instruct apiBase: https://api.scaleway.ai/v1/ apiKey: ###SCW_SECRET_KEY### defaultCompletionOptions: maxTokens: 8000 - contextLength: 50000 + contextLength: 100000 roles: - autocomplete - name: Embeddings Model - Scaleway @@ -127,8 +127,8 @@ Alternatively, a `config.json` file can be used with the following format. Note { "models": [ { - "model": "devstral-small-2505", - "title": "Devstral - Scaleway", + "model": "qwen3-235b-a22b-instruct-2507", + "title": "Qwen3 235B - Scaleway", "provider": "openai", "apiKey": "###SCW_SECRET_KEY###" } @@ -139,7 +139,7 @@ Alternatively, a `config.json` file can be used with the following format. Note "apiKey": "###SCW_SECRET_KEY###" }, "tabAutocompleteModel": { - "model": "qwen2.5-coder-32b-instruct", + "model": "qwen3-coder-30b-a3b-instruct", "title": "Autocomplete - Scaleway", "provider": "openai", "apiKey": "###SCW_SECRET_KEY###" From 585fb20060771cc8f6e82d796c0d683a04114a27 Mon Sep 17 00:00:00 2001 From: fpagny Date: Fri, 10 Oct 2025 16:57:18 +0200 Subject: [PATCH 2/4] feat(genapi): update IntelliJ Continue suggested models --- .../adding-ai-to-intellij-using-continue.mdx | 22 +++++++++---------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/pages/generative-apis/reference-content/adding-ai-to-intellij-using-continue.mdx b/pages/generative-apis/reference-content/adding-ai-to-intellij-using-continue.mdx index 8322949f3a..caa75f6b99 100644 --- a/pages/generative-apis/reference-content/adding-ai-to-intellij-using-continue.mdx +++ b/pages/generative-apis/reference-content/adding-ai-to-intellij-using-continue.mdx @@ -56,21 +56,21 @@ To link Continue with Scaleway’s Generative APIs, you need to configure the se - If you have already configured a **Local Assistant**, click **Local Assistant** then click the **wheel icon** to open your existing `config.yaml` - Otherwise, create a `config.yaml` file inside your `.continue` directory. - Add the following configuration to enable Scaleway's Generative API. This configuration uses three different models for each task: - - `devstral-small-2505` for agentic workflows through a chat interface - - `qwen2.5-coder-32b-instruct` for autocompletion when editing a file + - `qwen3-235b-a22b-instruct-2507` for agentic workflows through a chat interface (`qwen3-coder-30b-a3b-instruct` can also be used) + - `qwen3-coder-30b-a3b-instruct` for autocompletion when editing a file - `bge-multilingual-gemma2` for embedding and retrieving code context ```yaml name: Continue Config version: 0.0.1 models: - - name: Devstral - Scaleway + - name: Qwen3 235B - Scaleway provider: openai - model: devstral-small-2505 + model: qwen3-235b-a22b-instruct-2507 apiBase: https://api.scaleway.ai/v1/ apiKey: ###SCW_SECRET_KEY### defaultCompletionOptions: - maxTokens: 8000 - contextLength: 50000 + maxTokens: 4000 + contextLength: 40000 roles: - chat - apply @@ -80,12 +80,12 @@ To link Continue with Scaleway’s Generative APIs, you need to configure the se - tool_use - name: Autocomplete - Scaleway provider: openai - model: qwen2.5-coder-32b-instruct + model: qwen3-coder-30b-a3b-instruct apiBase: https://api.scaleway.ai/v1/ apiKey: ###SCW_SECRET_KEY### defaultCompletionOptions: maxTokens: 8000 - contextLength: 50000 + contextLength: 100000 roles: - autocomplete - name: Embeddings Model - Scaleway @@ -127,8 +127,8 @@ Alternatively, a `config.json` file can be used with the following format. Note { "models": [ { - "model": "devstral-small-2505", - "title": "Devstral - Scaleway", + "model": "qwen3-235b-a22b-instruct-2507", + "title": "Qwen3 235B - Scaleway", "provider": "openai", "apiKey": "###SCW_SECRET_KEY###" } @@ -139,7 +139,7 @@ Alternatively, a `config.json` file can be used with the following format. Note "apiKey": "###SCW_SECRET_KEY###" }, "tabAutocompleteModel": { - "model": "qwen2.5-coder-32b-instruct", + "model": "qwen3-coder-30b-a3b-instruct", "title": "Autocomplete - Scaleway", "provider": "openai", "apiKey": "###SCW_SECRET_KEY###" From 10f115f379fa2cc83e0d54fddceb20e9eda557c9 Mon Sep 17 00:00:00 2001 From: fpagny Date: Fri, 10 Oct 2025 17:01:21 +0200 Subject: [PATCH 3/4] feat(genapi): update Zed integration with Qwen3 Coder --- .../reference-content/adding-ai-to-zed-ide.mdx | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/pages/generative-apis/reference-content/adding-ai-to-zed-ide.mdx b/pages/generative-apis/reference-content/adding-ai-to-zed-ide.mdx index 864b3886fe..6c33bfe7bc 100644 --- a/pages/generative-apis/reference-content/adding-ai-to-zed-ide.mdx +++ b/pages/generative-apis/reference-content/adding-ai-to-zed-ide.mdx @@ -28,8 +28,8 @@ Zed is an IDE (Integrated Development Environment) including AI coding assistanc "api_url": "https://api.scaleway.ai/v1", "available_models": [ { - "name": "devstral-small-2505", - "display_name": "Devstral Small - Scaleway", + "name": "qwen3-coder-30b-a3b-instruct", + "display_name": "Qwen3 Coder - Scaleway", "max_tokens": 128000 } ], @@ -40,13 +40,13 @@ Zed is an IDE (Integrated Development Environment) including AI coding assistanc "default_profile": "write", "default_model": { "provider": "openai", - "model": "devstral-small-2505" + "model": "qwen3-coder-30b-a3b-instruct" } } } ``` - This configuration will add a `devstral-small-2505` Scaleway hosted model available with the Zed `openai` provider, and use it as default model. + This configuration will add a `qwen3-coder-30b-a3b-instruct` Scaleway hosted model available with the Zed `openai` provider, and use it as default model. 2. Open AI Assistant configuration by either using the command palette and typing `assistant: show configuration` or clicking on the bottom right **Assistant Panel** button and then **Assistant menu** in top right and finally **Configure**. @@ -55,5 +55,5 @@ Zed is an IDE (Integrated Development Environment) including AI coding assistanc Your key will be deleted if you restart Zed. To store it permanently, you can set up your Scaleway secret key as `OPENAI_API_KEY` environment variable and restart Zed. -4. Your setup is complete. If you open a new chat and select the `Devstral Small - Scaleway` model, you can send text and see the model using local tools to perform actions such as file reading or editing. You can review changes before they are applied and also add custom remote tools using [Custom MCP Servers](https://zed.dev/docs/ai/mcp). +4. Your setup is complete. If you open a new chat and select the `Qwen3 Coder - Scaleway` model, you can send text and see the model using local tools to perform actions such as file reading or editing. You can review changes before they are applied and also add custom remote tools using [Custom MCP Servers](https://zed.dev/docs/ai/mcp). Additionally, you can use the [Inline Assist](https://zed.dev/docs/ai/inline-assistant) feature when editing your code. From 70bcc35a4a992dff199b9e7fb1f82ca9df89a05f Mon Sep 17 00:00:00 2001 From: fpagny Date: Fri, 10 Oct 2025 17:23:34 +0200 Subject: [PATCH 4/4] feat(genapi): update troubleshooting Remove duplicated content, and add links to IDEs configuration for maximum context window reached. --- .../troubleshooting/fixing-common-issues.mdx | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) diff --git a/pages/generative-apis/troubleshooting/fixing-common-issues.mdx b/pages/generative-apis/troubleshooting/fixing-common-issues.mdx index ed7976a68b..f5db578f87 100644 --- a/pages/generative-apis/troubleshooting/fixing-common-issues.mdx +++ b/pages/generative-apis/troubleshooting/fixing-common-issues.mdx @@ -17,6 +17,7 @@ Below are common issues that you may encounter when using Generative APIs, their ### Solution - Reduce your input size below what is [supported by the model](/generative-apis/reference-content/supported-models/). + - If you are using a third party tool such as IDEs, you should edit their configuration to set an appropriate maximum context window for the model. More information for [VS Code (Continue)](/generative-apis/reference-content/adding-ai-to-vscode-using-continue/#configure-continue-through-a-configuration-file), [IntelliJ (Continue)](/generative-apis/reference-content/adding-ai-to-intellij-using-continue/#configure-continue-through-configuration-file) and [Zed](/generative-apis/reference-content/adding-ai-to-zed-ide/). - Use a model supporting longer context window values. - Use [Managed Inference](/managed-inference/), where the context window can be increased for [several configurations with additional GPU vRAM](/managed-inference/reference-content/supported-models/). For instance, `llama-3.3-70b-instruct` model in `fp8` quantization can be served with: - `15k` tokens context window on `H100` Instances @@ -51,20 +52,6 @@ Below are common issues that you may encounter when using Generative APIs, their ## 416: Range Not Satisfiable - max_completion_tokens is limited for this model -### Cause -- You provided a value for `max_completion_tokens` that is too high and not supported by the model you are using. - -### Solution -- Remove `max_completion_tokens` field from your request or client library, or reduce its value below what is [supported by the model](https://www.scaleway.com/en/docs/generative-apis/reference-content/supported-models/). - - As an example, when using the [init_chat_model from Langchain](https://python.langchain.com/api_reference/_modules/langchain/chat_models/base.html#init_chat_model), you should edit the `max_tokens` value in the following configuration: - ```python - llm = init_chat_model("llama-3.3-70b-instruct", max_tokens="8000", model_provider="openai", base_url="https://api.scaleway.ai/v1", temperature=0.7) - ``` -- Use a model supporting higher `max_completion_tokens` value. -- Use [Managed Inference](/managed-inference/), where these limits on completion tokens do not apply (your completion tokens amount will still be limited by the maximum context window supported by the model). - -## 416: Range Not Satisfiable - max_completion_tokens is limited for this model - ### Cause - You provided a value for `max_completion_tokens` which is too high, and not supported by the model you are using.