From e190006ff8f40afd72020917326105a89d0d2d7f Mon Sep 17 00:00:00 2001 From: fpagny Date: Fri, 14 Feb 2025 18:41:49 +0100 Subject: [PATCH 1/3] Update supported-models.mdx Add image size limitations for Pixtral model. --- .../reference-content/supported-models.mdx | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/pages/generative-apis/reference-content/supported-models.mdx b/pages/generative-apis/reference-content/supported-models.mdx index 42803b3b37..f82df8c192 100644 --- a/pages/generative-apis/reference-content/supported-models.mdx +++ b/pages/generative-apis/reference-content/supported-models.mdx @@ -15,7 +15,7 @@ dates: This service is free while in beta. Starting March 12th 2025, Generative APIs will become Generally Available, and the [following pricing](https://www.scaleway.com/en/pricing/model-as-a-service/#generative-apis) will apply along with exsting [Specific terms and conditions](https://www.scaleway.com/en/contracts/). -Our [Chat API](/generative-apis/how-to/query-language-models) has built-in support for the most popular instruct models. +Our API supports the most popular models for [Chat](/generative-apis/how-to/query-language-models), [Vision](/generative-apis/how-to/query-vision-models/) and [Embeddings](/generative-apis/how-to/query-embedding-models/) ## Chat models @@ -25,18 +25,22 @@ Our [Chat API](/generative-apis/how-to/query-language-models) has built-in suppo | Meta | `llama-3.1-70b-instruct` | 128k | 4096 | [Llama 3.1 Community](https://llama.meta.com/llama3_1/license/) | [HF](https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct) | | Meta | `llama-3.1-8b-instruct` | 128k | 16384 | [Llama 3.1 Community](https://llama.meta.com/llama3_1/license/) | [HF](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) | | Mistral | `mistral-nemo-instruct-2407` | 128k | 8192 | [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) | [HF](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) | -| Mistral | `pixtral-12b-2409` | 128k | 4096 | [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) | [HF](https://huggingface.co/mistralai/Pixtral-12B-2409) | | Qwen | `qwen2.5-coder-32b-instruct` | 128k | 8192 | [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) | [HF](https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct) | | DeepSeek (Preview) | `deepseek-r1` | 20k | 4096 | [MIT](https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/mit.md) | [HF](https://huggingface.co/deepseek-ai/DeepSeek-R1) | | DeepSeek (Preview) | `deepseek-r1-distill-llama-70b` | 32k | 4096 | [MIT](https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/mit.md) | [HF](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B) | - If you are unsure which chat model to use, we currently recommend Llama 3.1 8B Instruct (`llama-3.1-8b-instruct`) to get started. +## Vision models + +| Provider | Model string | Context window (Tokens) | Maximum output (Tokens)| License | Model card | +|-----------------|-----------------|-----------------|-----------------|-----------------|-----------------| +| Mistral | `pixtral-12b-2409` | 128k | 4096 | [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) | [HF](https://huggingface.co/mistralai/Pixtral-12B-2409) | + - Missing a model? [Tell us what you would like to see here](https://feature-request.scaleway.com/?tags=ai-services), and we might add it shortly. + Images size are limited to 32 million pixels (eg. around 8096x4048 resolution). Note that images above 1024x1024 are supported, but will be downscaled automatically to fit within this limits (image ratio and proportions will be preserved). ## Embedding models From 5e6ff3b47c818dce21c48e6bce2bd7834e4c07e2 Mon Sep 17 00:00:00 2001 From: Benedikt Rollik Date: Mon, 17 Feb 2025 10:32:55 +0100 Subject: [PATCH 2/3] Update pages/generative-apis/reference-content/supported-models.mdx Co-authored-by: Jessica <113192637+jcirinosclwy@users.noreply.github.com> --- pages/generative-apis/reference-content/supported-models.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pages/generative-apis/reference-content/supported-models.mdx b/pages/generative-apis/reference-content/supported-models.mdx index f82df8c192..c7122f4884 100644 --- a/pages/generative-apis/reference-content/supported-models.mdx +++ b/pages/generative-apis/reference-content/supported-models.mdx @@ -15,7 +15,7 @@ dates: This service is free while in beta. Starting March 12th 2025, Generative APIs will become Generally Available, and the [following pricing](https://www.scaleway.com/en/pricing/model-as-a-service/#generative-apis) will apply along with exsting [Specific terms and conditions](https://www.scaleway.com/en/contracts/). -Our API supports the most popular models for [Chat](/generative-apis/how-to/query-language-models), [Vision](/generative-apis/how-to/query-vision-models/) and [Embeddings](/generative-apis/how-to/query-embedding-models/) +Our API supports the most popular models for [Chat](/generative-apis/how-to/query-language-models), [Vision](/generative-apis/how-to/query-vision-models/) and [Embeddings](/generative-apis/how-to/query-embedding-models/). ## Chat models From 60a8a9403544b6b4ed9b0a6d3b3032f9ce57ea06 Mon Sep 17 00:00:00 2001 From: Benedikt Rollik Date: Mon, 17 Feb 2025 10:41:12 +0100 Subject: [PATCH 3/3] Apply suggestions from code review Co-authored-by: Jessica <113192637+jcirinosclwy@users.noreply.github.com> --- pages/generative-apis/reference-content/supported-models.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pages/generative-apis/reference-content/supported-models.mdx b/pages/generative-apis/reference-content/supported-models.mdx index c7122f4884..14fc29a9f9 100644 --- a/pages/generative-apis/reference-content/supported-models.mdx +++ b/pages/generative-apis/reference-content/supported-models.mdx @@ -40,7 +40,7 @@ Our API supports the most popular models for [Chat](/generative-apis/how-to/quer | Mistral | `pixtral-12b-2409` | 128k | 4096 | [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) | [HF](https://huggingface.co/mistralai/Pixtral-12B-2409) | - Images size are limited to 32 million pixels (eg. around 8096x4048 resolution). Note that images above 1024x1024 are supported, but will be downscaled automatically to fit within this limits (image ratio and proportions will be preserved). + Image sizes are limited to 32 million pixels (e.g., a resolution of about 8096 x 4048). Images with a resolution higher than 1024 x 1024 are supported, but automatically downscaled to fit these limitations (image ratio and proportions will be preserved). ## Embedding models