Conversation
Adding Llama 3.3
Fixes typo
…ocker#53) * add Granite Docling model card with description and characteristics * Apply suggestion from @gemini-code-assist[bot] Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Apply suggestion from @gemini-code-assist[bot] Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Apply suggestion from @gemini-code-assist[bot] Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Apply suggestion from @gemini-code-assist[bot] Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Apply suggestion from @gemini-code-assist[bot] Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> --------- Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
* add gpt-oss-safeguard * Update ai/gpt-oss-safeguard.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> --------- Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>
* add snowflakw-arctic-embed-l-v2-vllm * Update ai/snowflakw-arctic-embed-l-v2-vllm.md Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> --------- Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
* add qwen3-embedding * Update ai/qwen3-embedding.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * Update ai/qwen3-embedding-vllm.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * Update ai/qwen3-embedding-vllm.md Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Update ai/qwen3-embedding-vllm.md Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> * remove unnecessary blank lines in qwen3-embedding and qwen3-embedding-vllm documentation --------- Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
* ai: add qwen3-reranker for vLLM Signed-off-by: Dorin Geman <dorin.geman@docker.com> * ai: add qwen3-reranker Signed-off-by: Dorin Geman <dorin.geman@docker.com> --------- Signed-off-by: Dorin Geman <dorin.geman@docker.com>
* fix endpoint URL for snowflakw-arctic-embed-l-v2-vllm model * add Ministral 3 model card with detailed specifications and use cases
Signed-off-by: Dorin Geman <dorin.geman@docker.com>
ai: add qwen3-coder-next
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
GitHub Actions only reads workflows from the repository root's .github/workflows/ directory. Updated tool paths to reflect the new model-cards/ subtree prefix. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Model cards from the imported repository live under model-cards/ai/, not model-cards/. Update the generator workflow and agent config to write to the correct location. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
There was a problem hiding this comment.
Sorry @doringeman, your pull request is larger than the review limit of 150000 diff characters
Summary of ChangesHello @doringeman, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request integrates the external Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Changelog
Ignored Files
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
All content now lives under model-cards/ in the model-runner monorepo. See docker/model-runner#650. Signed-off-by: Dorin Geman <dorin.geman@docker.com>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request is a significant contribution, merging the standalone model-cards repository into this one. It adds a wealth of model documentation and a new CLI tool. My review focuses on the newly added markdown files, where I've identified several areas for improvement to ensure consistency, correctness, and maintainability. The key issues include broken image links due to absolute URLs pointing to the old repository, inconsistencies in command examples, and some duplicated content in model variant tables. Addressing these points will greatly improve the quality and usability of the documentation.
| ## 🚀 Models Overview | ||
|
|
||
| ### DeepCoder Preview | ||
|  |
There was a problem hiding this comment.
The logo URLs in this README and other model card files are absolute links to the old docker/model-cards GitHub repository. After this merge, these links will be fragile and likely break. They should be updated to relative paths to point to the new model-cards/logos/ directory within this repository.
For example,  should become .
Furthermore, some of the logos referenced (like for DeepCoder Preview and GPT-OSS Safeguard) are not present in the model-cards/logos directory in this pull request, which will result in broken images.
| 📌 **Description:** | ||
| 24B multimodal instruction model by Mistral AI, tuned for accuracy, tool use & fewer repeats. | ||
|
|
||
| 📂 **Model File:** [`ai/magistral-small-3.2.md`](ai/magistral-small-2506.md) |
There was a problem hiding this comment.
The link for the model file is broken. The link text ai/magistral-small-3.2.md is correct, but the target points to ai/magistral-small-2506.md, which does not exist.
| 📂 **Model File:** [`ai/magistral-small-3.2.md`](ai/magistral-small-2506.md) | |
| 📂 **Model File:** [`ai/magistral-small-3.2.md`](ai/magistral-small-3.2.md) |
| 24B multimodal instruction model by Mistral AI, tuned for accuracy, tool use & fewer repeats. | ||
|
|
||
| 📂 **Model File:** [`ai/magistral-small-3.2.md`](ai/magistral-small-2506.md) | ||
| 🐳 **Docker Hub:** [`docker.io/ai/magistral-small-3.2`](https://hub.docker.com/r/ai/magistral-small-2506) |
There was a problem hiding this comment.
The Docker Hub link for this model appears to be incorrect. The link target .../magistral-small-2506 does not match the model name magistral-small-3.2.
| 🐳 **Docker Hub:** [`docker.io/ai/magistral-small-3.2`](https://hub.docker.com/r/ai/magistral-small-2506) | |
| 🐳 **Docker Hub:** [`docker.io/ai/magistral-small-3.2`](https://hub.docker.com/r/ai/magistral-small-3.2) |
| First, pull the model: | ||
|
|
||
| ```bash | ||
| docker model pull {model_name} |
There was a problem hiding this comment.
The command uses a placeholder {model_name}. It should be replaced with the actual model name for the command to be runnable.
| docker model pull {model_name} | |
| docker model pull ai/granite-embedding-multilingual |
| Then run the model: | ||
|
|
||
| ```bash | ||
| docker model run {model_name} |
There was a problem hiding this comment.
The command uses a placeholder {model_name}. It should be replaced with the actual model name for the command to be runnable.
| docker model run {model_name} | |
| docker model run ai/granite-embedding-multilingual |
| --- | ||
|
|
||
| ### SmolLM 2 | ||
|  |
There was a problem hiding this comment.
There's a typo in the logo URL: hugginfface should be huggingface. This typo is also present in the filename itself (hugginfface-120x-hub@2x.svg). Please correct the link and also consider renaming the file model-cards/logos/hugginfface-120x-hub@2x.svg to huggingface-120x-hub@2x.svg for consistency. This issue is also present on line 340.
|  | |
|  |
| | Model variant | Parameters | Quantization | Context window | VRAM¹ | Size | | ||
| |---------------|------------|--------------|----------------|------|-------| | ||
| | `ai/deepcoder-preview:latest`<br><br>`ai/deepcoder-preview:14B-Q4_K_M` | 14B | IQ2_XXS/Q4_K_M | 131K tokens | 9.36 GiB | 8.37 GB | | ||
| | `ai/deepcoder-preview:14B-Q4_K_M` | 14B | IQ2_XXS/Q4_K_M | 131K tokens | 9.36 GiB | 8.37 GB | |
There was a problem hiding this comment.
This line is a duplicate of the 14B-Q4_K_M variant already listed in the line above. Please remove this redundant entry to avoid confusion. This issue is present in several other model card files as well.
| ## Use this AI model with Docker Model Runner | ||
|
|
||
| ```bash | ||
| docker model run deepseek-v3.2-vllm |
There was a problem hiding this comment.
The model name deepseek-v3.2-vllm in the command is inconsistent with the model card's filename (deepseek3.2.md) and the ai/ namespacing convention used in most other model cards. For consistency and clarity, please consider using ai/deepseek3.2. This inconsistency is also present in several other new model cards.
| docker model run deepseek-v3.2-vllm | |
| docker model run ai/deepseek3.2 |
| |-----------------------|----------------------------------------------------------------------------------------------------------------------| | ||
| | **Provider** | IBM (Granite Embedding Team) | | ||
| | **Architecture** | Encoder‑only transformer, XLM‑RoBERTa‑like bi‑encoder | | ||
| | **Cutoff date** | Released December 18, 2024:contentReference | |
There was a problem hiding this comment.
The cutoff date contains what appears to be a placeholder or error (:contentReference). Please remove it or replace it with the correct information.
| | **Cutoff date** | Released December 18, 2024:contentReference | | |
| | **Cutoff date** | Released December 18, 2024 | |
| - **AI assistance on edge devices**, Running chatbots and virtual assistants with minimal latency on low-power * hardware. | ||
| - **Code assistance** , Writing, debugging, and optimizing code on mobile or embedded systems. | ||
| - **Content generation** ,Drafting emails, summaries, and creative content on lightweight devices. | ||
| - **Low-power AI for smart gadgets**, Enhancing voice assistants on wearables and IoT devices. | ||
| - **Edge-based data processing**, Summarizing and analyzing data locally for security and efficiency. |
There was a problem hiding this comment.
This list has some minor formatting issues: a stray asterisk on the first line, and inconsistent spacing around commas on subsequent lines. The suggested change improves readability and consistency.
| - **AI assistance on edge devices**, Running chatbots and virtual assistants with minimal latency on low-power * hardware. | |
| - **Code assistance** , Writing, debugging, and optimizing code on mobile or embedded systems. | |
| - **Content generation** ,Drafting emails, summaries, and creative content on lightweight devices. | |
| - **Low-power AI for smart gadgets**, Enhancing voice assistants on wearables and IoT devices. | |
| - **Edge-based data processing**, Summarizing and analyzing data locally for security and efficiency. | |
| - **AI assistance on edge devices**: Running chatbots and virtual assistants with minimal latency on low-power hardware. | |
| - **Code assistance**: Writing, debugging, and optimizing code on mobile or embedded systems. | |
| - **Content generation**: Drafting emails, summaries, and creative content on lightweight devices. | |
| - **Low-power AI for smart gadgets**: Enhancing voice assistants on wearables and IoT devices. | |
| - **Edge-based data processing**: Summarizing and analyzing data locally for security and efficiency. |
All content now lives under model-cards/ in the model-runner monorepo. See docker/model-runner#650. Signed-off-by: Dorin Geman <dorin.geman@docker.com>
Consolidates the standalone docker/model-cards repository into
model-cards/usinggit subtree add, preserving full commit history (110 commits).Relocates
publish-model-card.ymlworkflow to top-level.github/workflows/(GitHub Actions requirement) and updates tool paths.Updates
generate-model-cardworkflow and agent config to write tomodel-cards/ai/instead ofmodel-cards/.