diff --git a/docs/cli/Guides/Solutions/tgwui.md b/docs/cli/Guides/Solutions/tgwui.md deleted file mode 100644 index ca7c461b..00000000 --- a/docs/cli/Guides/Solutions/tgwui.md +++ /dev/null @@ -1,8 +0,0 @@ ---- -id: "tgwui" -title: "Text Generation WebUI" -slug: "/guides/solutions/tgwui" -sidebar_position: 1 -displayed_sidebar: null -unlisted: true ---- \ No newline at end of file diff --git a/docs/cli/Guides/collaboration.md b/docs/cli/Guides/collaboration.md index 5c593118..47e5a7d8 100644 --- a/docs/cli/Guides/collaboration.md +++ b/docs/cli/Guides/collaboration.md @@ -1,6 +1,6 @@ --- id: "collaboration" -title: "Confidential Collaboration" +title: "Two-party Collaboration" slug: "/guides/collaboration" displayed_sidebar: null unlisted: true diff --git a/docs/cli/Guides/Solutions/comfyui.md b/docs/cli/Guides/comfyui.md similarity index 99% rename from docs/cli/Guides/Solutions/comfyui.md rename to docs/cli/Guides/comfyui.md index aa1cab11..ddc23efb 100644 --- a/docs/cli/Guides/Solutions/comfyui.md +++ b/docs/cli/Guides/comfyui.md @@ -1,8 +1,8 @@ --- id: "comfyui" -title: "ComfyUI" +title: "Custom ComfyUI Workflow" slug: "/guides/solutions/comfyui" -sidebar_position: 2 +sidebar_position: 4 --- import Tabs from '@theme/Tabs'; diff --git a/docs/cli/Guides/confidential-fine-tuning.md b/docs/cli/Guides/multi-party-collab.md similarity index 83% rename from docs/cli/Guides/confidential-fine-tuning.md rename to docs/cli/Guides/multi-party-collab.md index f68d7f63..b9176b09 100644 --- a/docs/cli/Guides/confidential-fine-tuning.md +++ b/docs/cli/Guides/multi-party-collab.md @@ -1,13 +1,13 @@ --- id: "fine-tune" -title: "Confidential Collaboration" -slug: "/guides/fine-tune" -sidebar_position: 3 +title: "Multi-Party Collaboration" +slug: "/guides/multi-party-collab" +sidebar_position: 7 --- Super Protocol enables independent parties to jointly compute over their private inputs without revealing those inputs to one another. -This guide describes a scenario of confidential collaboration on Super Protocol. It uses fine-tuning of a pre-trained AI model as an example, but the general principle presented here may be applied to other cases. +This guide describes a scenario of a multi-party confidential collaboration on Super Protocol. It uses fine-tuning of a pre-trained AI model as an example, but the general principle presented here may be applied to other cases. The scenario involves three parties: @@ -41,11 +41,11 @@ sequenceDiagram ```
-Carol builds a solution—a Docker image containing her training engine and script ([1](/cli/guides/fine-tune#carol-1-build-a-solution)). She uploads the solution using SPCTL ([2](/cli/guides/fine-tune#carol-2-upload-the-solution)) and grants Alice and Bob access for verification ([3](/cli/guides/fine-tune#carol-3-send-the-solution-to-alice-and-bob)). +Carol builds a solution—a Docker image containing her training engine and script ([1](/cli/guides/multi-party-collab#carol-1-build-a-solution)). She uploads the solution using SPCTL ([2](/cli/guides/multi-party-collab#carol-2-upload-the-solution)) and grants Alice and Bob access for verification ([3](/cli/guides/multi-party-collab#carol-3-send-the-solution-to-alice-and-bob)). -Alice and Bob download the solution ([4](/cli/guides/fine-tune#alice-and-bob-4-download-the-solution)) and verify that it is safe to process their data ([5](/cli/guides/fine-tune#alice-and-bob-5-verify-the-solution)). +Alice and Bob download the solution ([4](/cli/guides/multi-party-collab#alice-and-bob-4-download-the-solution)) and verify that it is safe to process their data ([5](/cli/guides/multi-party-collab#alice-and-bob-5-verify-the-solution)). -If Carol cannot share the code with Alice and Bob, and a third-party auditor is involved, the workflow will differ slightly. After uploading, Carol must also create a solution offer on Super Protocol Marketplace (similar to Bob's [Step 8](/cli/guides/fine-tune#bob-8-create-an-offer)). +If Carol cannot share the code with Alice and Bob, and a third-party auditor is involved, the workflow will differ slightly. After uploading, Carol must also create a solution offer on Super Protocol Marketplace (similar to Bob's [Step 8](/cli/guides/multi-party-collab#bob-8-create-an-offer)). **Prepare the data**: @@ -64,9 +64,9 @@ sequenceDiagram ```
-Alice uploads her model ([6](/cli/guides/fine-tune#alice-6-upload-the-model)) and Bob uploads his dataset ([7](/cli/guides/fine-tune#bob-7-upload-the-dataset)) to remote storage using SPCTL. Files are automatically encrypted during upload, and only the uploader holds the key. +Alice uploads her model ([6](/cli/guides/multi-party-collab#alice-6-upload-the-model)) and Bob uploads his dataset ([7](/cli/guides/multi-party-collab#bob-7-upload-the-dataset)) to remote storage using SPCTL. Files are automatically encrypted during upload, and only the uploader holds the key. -Bob creates an offer on the Marketplace ([8](/cli/guides/fine-tune#bob-8-create-an-offer)). The offer requires Bob's manual approval for use. He shares the offer's IDs with Alice. +Bob creates an offer on the Marketplace ([8](/cli/guides/multi-party-collab#bob-8-create-an-offer)). The offer requires Bob's manual approval for use. He shares the offer's IDs with Alice. **Execute**: @@ -92,13 +92,13 @@ sequenceDiagram ```
-Alice places an order on Super Protocol ([9](/cli/guides/fine-tune#alice-9-place-an-order)), adding the solution, her model, and Bob's offer. The order does not proceed automatically and remains `Blocked`. +Alice places an order on Super Protocol ([9](/cli/guides/multi-party-collab#alice-9-place-an-order)), adding the solution, her model, and Bob's offer. The order does not proceed automatically and remains `Blocked`. -Bob manually approves the usage of his dataset for the image with a specific hash ([10](/cli/guides/fine-tune#bob-10-complete-the-data-suborder)). If this hash matches the actual solution hash, the CVM begins to process the order. If the hashes do not match, the order will be terminated with an error. +Bob manually approves the usage of his dataset for the image with a specific hash ([10](/cli/guides/multi-party-collab#bob-10-complete-the-data-suborder)). If this hash matches the actual solution hash, the CVM begins to process the order. If the hashes do not match, the order will be terminated with an error. -When the order is complete, Alice downloads the result ([11](/cli/guides/fine-tune#alice-11-download-the-order-results)). All the data within the TEE (solution, AI model, dataset, order results, etc.) is automatically deleted. +When the order is complete, Alice downloads the result ([11](/cli/guides/multi-party-collab#alice-11-download-the-order-results)). All the data within the TEE (solution, AI model, dataset, order results, etc.) is automatically deleted. -Both Alice and Bob can retrieve the order report ([12](/cli/guides/fine-tune#alice-and-bob-12-get-the-order-report)) that confirms the authenticity of the entire trusted setup. +Both Alice and Bob can retrieve the order report ([12](/cli/guides/multi-party-collab#alice-and-bob-12-get-the-order-report)) that confirms the authenticity of the entire trusted setup. ## Prerequisites @@ -138,7 +138,7 @@ Always use absolute paths, such as `/sp/...`. ::: -You may either include the training scripts in the image or upload them separately using SPCTL. In this case, Alice will need to attach the uploaded scripts to the order as `--data` at [Step 9](/cli/guides/fine-tune#alice-9-place-an-order). +You may either include the training scripts in the image or upload them separately using SPCTL. In this case, Alice will need to attach the uploaded scripts to the order as `--data` at [Step 9](/cli/guides/multi-party-collab#alice-9-place-an-order). You can find several Dockerfile examples in the [Super-Protocol/solutions](https://github.com/Super-Protocol/solutions) GitHub repository. diff --git a/docs/cli/Guides/provider-tools.md b/docs/cli/Guides/provider-tools.md index 3b18dd60..8c6c7f79 100644 --- a/docs/cli/Guides/provider-tools.md +++ b/docs/cli/Guides/provider-tools.md @@ -1,6 +1,6 @@ --- id: "provider-tools" -title: "Configure Provider Tools" +title: "Provider Tools" slug: "/guides/provider-tools" sidebar_position: 1 --- diff --git a/docs/cli/Guides/quick-guide.md b/docs/cli/Guides/quick-guide.md index 0092c4a8..89d14492 100644 --- a/docs/cli/Guides/quick-guide.md +++ b/docs/cli/Guides/quick-guide.md @@ -1,11 +1,11 @@ --- id: "quick-guide" -title: "Quick Deployment Guide" +title: "Deploy Your App" slug: "/guides/quick-guide" sidebar_position: 2 --- -This quick guide provides instructions on deploying a solution and data on Super Protocol. Its purpose is to introduce you to the logic and sequence of the CLI commands. +This quick guide provides instructions on deploying your own solution and data on Super Protocol. Its purpose is to introduce you to the logic and sequence of the CLI commands. ## Prerequisites diff --git a/docs/cli/Guides/tgwui.md b/docs/cli/Guides/tgwui.md new file mode 100644 index 00000000..e12f58ef --- /dev/null +++ b/docs/cli/Guides/tgwui.md @@ -0,0 +1,234 @@ +--- +id: "tgwui" +title: "TGWUI and ComfyUI With Tunnels" +slug: "/guides/solutions/tgwui" +sidebar_position: 3 +--- + +import Tabs from '@theme/Tabs'; +import TabItem from '@theme/TabItem'; + +This guide provides step-by-step instructions for uploading and deploying an AI model on Super Protocol using Text Generation Web UI and ComfyUI, both already available in the Marketplace. However, the general workflow described here can be applied to any solution, whether new or existing. + +## Prerequisites + +- [SPCTL](https://docs.develop.superprotocol.com/cli/) +- BNB and SPPI tokens (opBNB) to pay for transactions and orders + +## 1. Prepare + +Ensure your model meets the Super Protocol requirements: + +1.1. Your model must belong to a category supported by one of the engines. + +**Text Generation Web UI**: + +- Text Generation +- Text Classification +- Translation +- Text2Text Generation + +**ComfyUI**: + +- Image Classification +- Object Detection +- Image Segmentation +- Text-to-Image +- Image-to-Text +- Image-to-Image +- Image-to-Video +- Video Classification +- Text-to-Video +- Mask Generation + +If you plan to deploy a ComfyUI workflow with custom nodes, [prepare the files](/cli/guides/solutions/comfyui) before proceeding to the next step. For security reasons, you cannot upload custom nodes directly to a deployed ComfyUI. + +1.2. Due to [testnet limitations](/marketplace/limitations), the total size of model files should not exceed 13 GB. Support for bigger models will be available in the future. + +## 2. Upload the model + +Upload the model: + +```shell +./spctl files upload \ + --output model.resource.json \ + --use-addon +``` + +Replace `` with the path to the dataset directory, for example: + +```shell +./spctl files upload ~/Downloads/models/SmolLM2-1.7B \ + --output model.resource.json \ + --use-addon +``` + +## 3. Deploy tunnels + +3.1. Place an order to deploy a [confidential tunnel](/fundamentals/tunnels): + +```shell +./spctl workflows create --tee 7 --solution 19 +``` + +3.2. Wait for the order to be created, and find the tunnel order ID in the output, for example: + +```text +Workflow was created, TEE order id: ["273899"] +``` + +3.3. Check the order status: + +```shell +./spctl orders get +``` + +Replace `` with the tunnel order ID from the previous step. + +3.4. When the status is `Done`, download the result: + +```shell +./spctl orders download-result +``` + +3.5. Extract the downloaded `result.tar.gz`, open `output/result.json`, and find the domain address. For example: + +```json title="result.json" +"domain":"pret-tons-wade.superprotocol.io" +``` + +Your model's web UI will be available at this address. + +## 4. Prepare engine configuration files + +4.1. Open the SPCTL's `config.json` and find the `workflow.resultEncryption.key` property that contains the key used for decrypting workflow results; for example: `NapSrwQRz2tL9ZftJbi6DATpCDn0BRImpSStU9xZT/s=`. + +4.2. + + + Create a file named `engine-configuration-tgwui.json` and paste the following: + + ```json title="engine-configuration-tgwui.json" + { + "engine": { + "main_settings": { + "character": { + "name": "Superprotocol AI", + "context": "The following is a conversation with an AI Large Language Model. The AI has been trained to answer questions, provide recommendations, and help with decision making. The AI follows user requests. The AI thinks outside the box.", + "greeting": "How can I help you today?" + }, + "api": {}, + "mode": {} + }, + "model": { + "parameters": { + "temperature": 1, + "top_p": 1, + "top_k": 0, + "typical_p": 1 + }, + "parameters2": { + "min_p": 0.05, + "repetition_penalty": 1, + "frequency_penalty": 0, + "presence_penalty": 0 + } + }, + "model_loader": { + "loader_name": "Autodetect" + } + }, + "tunnels": { + "domain_settings": { + "provision_type": "Temporary Domain (on *.superprotocol.io)", + "tunnel_provisioner_order": { + "order_id": "", + "order_key": "" + } + } + } + } + ``` + + + Create a file named `engine-configuration-comfyui.json` and paste the following: + + ```json title="engine-configuration-comfyui.json" + { + "engine": { + "main_settings": { + "preview_method": "none", + "preview_size": 512 + } + }, + "tunnels": { + "domain_settings": { + "provision_type": "Temporary Domain (on *.superprotocol.io)", + "tunnel_provisioner_order": { + "order_id": "", + "order_key": "" + } + } + } + } + ``` + + + +In `tunnels.domain_settings.tunnel_provisioner_order`, set: + +- `order_id` to your tunnel order ID from Step 3.2 +- `order_key` to your encryption key from Step 4.1 + +Save and close the file. + +## 5. Deploy the model + +5.1. + + + Create the main order to deploy your uploaded model: + + ```shell + ./spctl workflows create --tee --solution 25 --solution-configuration ./engine-configuration-tgwui.json --data ./model.resource.json + ``` + + Replace `` with a selected compute offer. See available compute offer IDs on the [Marketplace](https://marketplace.superprotocol.com/). + + Note that `--solution 25` refers to [Text Generation Web UI with GPU support](https://marketplace.superprotocol.com/marketplace/models?offer=offerId%3D25). If you need the CPU version, use `--solution 26` instead. + + + Create the main order to deploy your uploaded model: + + ```shell + ./spctl workflows create --tee --solution 27 --solution-configuration ./engine-configuration-comfyui.json --data ./model.resource.json + ``` + + Replace `` with a selected compute offer. See available compute offer IDs on the [Marketplace](https://marketplace.superprotocol.com/). + + Note that `--solution 27` refers to [ComfyUI UI with GPU support](https://marketplace.superprotocol.com/marketplace/models?offer=offerId%3D27). If you need the CPU version, use `--solution 28` instead. + + + +5.2. Wait for the order to be created, and find the main order ID in the output, for example: + +```text +Workflow was created, TEE order id: ["273900"] +``` + +5.3. Deployment may take 15-20 minutes or more, depending on the model size and other parameters. Check the domain from Step 3.5 every few minutes until the UI is available. + +If you suspect something went wrong, check the order status: + +```shell +./spctl orders get +``` + +Replace `` with the main order ID from the previous step. + +The most important statuses (see the [full list](/fundamentals/orders#compute-order)): + +- **Processing**: The compute is executing the order inside a TEE. Your model is either already available or will be available soon. +- **In Queue**: The order is waiting for the compute to become available. This status appears only if the compute is overloaded with orders. If this status persists for a few minutes, place a new main order the same tunnel order and engine configuration but another compute offer. +- **Done**: The order is completed successfully and the model's UI is no longer available. +- **Error**: The order completed with an error. [Download the order results](/cli/commands/orders/download-result) to get more information about the error. \ No newline at end of file diff --git a/docs/cli/Guides/Solutions/unsloth.md b/docs/cli/Guides/unsloth.md similarity index 98% rename from docs/cli/Guides/Solutions/unsloth.md rename to docs/cli/Guides/unsloth.md index 02d0b996..24b46110 100644 --- a/docs/cli/Guides/Solutions/unsloth.md +++ b/docs/cli/Guides/unsloth.md @@ -1,13 +1,13 @@ --- id: "unsloth" -title: "Unsloth" +title: "Fine-Tuning With Unsloth" slug: "/guides/solutions/unsloth" -sidebar_position: 3 +sidebar_position: 5 --- This guide provides step-by-step instructions for fine-tuning an AI model using the Super Protocol packaging of [Unsloth](https://unsloth.ai/), an open-source framework for LLM fine-tuning and reinforcement learning. -The solution allows you to run fine-tuning within Super Protocol's Trusted Execution Environment (TEE). This provides enhanced security and privacy and enables a range of [confidential collaboration](https://docs.develop.superprotocol.com/cli/guides/fine-tune) scenarios. +The solution allows you to run fine-tuning within Super Protocol's Trusted Execution Environment (TEE). This provides enhanced security and privacy and enables a range of [confidential collaboration](https://docs.develop.superprotocol.com/cli/guides/multi-party-collab) scenarios. ## Prerequisites diff --git a/docs/cli/Guides/Solutions/vllm.md b/docs/cli/Guides/vllm.md similarity index 92% rename from docs/cli/Guides/Solutions/vllm.md rename to docs/cli/Guides/vllm.md index cda0ab5d..a833a22d 100644 --- a/docs/cli/Guides/Solutions/vllm.md +++ b/docs/cli/Guides/vllm.md @@ -1,13 +1,11 @@ --- id: "vllm" -title: "vLLM" +title: "Inference With vLLM" slug: "/guides/solutions/vllm" -sidebar_position: 4 +sidebar_position: 6 --- -This guide provides step-by-step instructions for running an AI model inference using the Super Protocol packaging of [vLLM](https://www.vllm.ai/), an inference and serving engine for LLMs. - -The solution allows you to run LLM inference within Super Protocol's Trusted Execution Environment (TEE). +This guide provides step-by-step instructions for running an AI model inference using the Super Protocol packaging of [vLLM](https://www.vllm.ai/), an inference and serving engine for LLMs. This solution allows you to run LLM inference within Super Protocol's Trusted Execution Environment (TEE). ## Prerequisites diff --git a/docs/fundamentals/orders.md b/docs/fundamentals/orders.md index 5bd30508..8532ed07 100644 --- a/docs/fundamentals/orders.md +++ b/docs/fundamentals/orders.md @@ -75,7 +75,8 @@ Statuses: - **New**: The order is waiting for the response from the compute provider. - **In Queue**: The order is waiting in the queue for the compute to become available. This status appears only if the compute is overloaded with orders. - **Processing**: The compute is executing the order inside a TEE. -- **Done**: The order is completed. +- **Done**: The order is completed successfully. +- **Error**: The order completed with an error. Note that the **Processing** and **Done** statuses may have different meanings depending on the usage scenario. For one-time orders, such as executing a Python script, **Processing** means that the machine is working with the solution and data. When this is over, the main order becomes **Done**. diff --git a/docs/guides/index.md b/docs/guides/index.md index dd581766..31dab108 100644 --- a/docs/guides/index.md +++ b/docs/guides/index.md @@ -23,7 +23,7 @@ sidebar_position: 0 | [Configure SPCTL](/cli) | How to set up SPCTL—a Super Protocol CLI tool. | | [Configure Provider Tools](/cli/guides/provider-tools) | How to set up Provider Tools—a Super Protocol CLI utility for registering providers and creating offers. | | [Quick Deployment Guide](/cli/guides/quick-guide) | Quick instructions on deploying a solution and data on Super Protocol. | -| [Confidential Collaboration](/cli/guides/fine-tune) | A scenario of confidential collaboration on Super Protocol. | +| [Confidential Collaboration](/cli/guides/multi-party-collab) | A scenario of confidential collaboration on Super Protocol. | ### Solutions diff --git a/src/.DS_Store b/src/.DS_Store deleted file mode 100644 index bdf7d604..00000000 Binary files a/src/.DS_Store and /dev/null differ diff --git a/src/pages/.DS_Store b/src/pages/.DS_Store deleted file mode 100644 index d86d37c8..00000000 Binary files a/src/pages/.DS_Store and /dev/null differ diff --git a/static/.DS_Store b/static/.DS_Store deleted file mode 100644 index 20fef11f..00000000 Binary files a/static/.DS_Store and /dev/null differ