# microsoft/fara **Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/microsoft-fara).** 5,901 stars · 571 forks · Python · MIT ## Links - GitHub: https://github.com/microsoft/fara - awesome-repositories: https://awesome-repositories.com/repository/microsoft-fara.md ## Topics `agent` `browser-use` `computer-use` `computer-use-agent` `cua` ## Description FARA is a visual computer-use agent model that controls a browser by predicting screen coordinates for clicking, typing, and scrolling, without relying on DOM or accessibility trees. It is designed to automate multi-step web tasks such as searching, form filling, booking, and shopping by reasoning over visual state and decomposing tasks into sequential actions. The model uses a compact 7-billion-parameter decoder-only transformer that can run on consumer GPUs for low-latency on-device inference, or be deployed as a managed endpoint on Azure Foundry for cloud-based inference without local infrastructure. It also supports self-hosted serving via vLLM, LM Studio, or Ollama, giving users full control over the inference environment. FARA includes a reproducible evaluation framework that runs agent benchmarks on 609 real, live web-browsing tasks with automatic retry handling for time-sensitive and error-prone scenarios. The framework provides standardized scoring rubrics to compare agent performance across different task descriptions and versions. ## Tags ### Part of an Awesome List - [Browser and Web Agents](https://awesome-repositories.com/f/awesome-lists/ai/browser-and-web-agents.md) — Controls a browser to complete multi-step web tasks through visual perception and coordinate-based actions. - [Agent Benchmarks](https://awesome-repositories.com/f/awesome-lists/ai/agent-benchmarks.md) — Provides a reproducible evaluation system for testing web-browsing agents across 609 tasks with live websites. ### Business & Productivity Software - [Visual Web Task Agents](https://awesome-repositories.com/f/business-productivity-software/web-task-automations/visual-web-task-agents.md) — Provides a visual computer-use agent that automates multi-step web tasks through pixel-level interface control. - [Web Task Automations](https://awesome-repositories.com/f/business-productivity-software/web-task-automations.md) — Automates multi-step web tasks like searching, form filling, booking, and shopping via visual interface control. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) ### Artificial Intelligence & ML - [Task Decomposition](https://awesome-repositories.com/f/artificial-intelligence-ml/agent-architectures/orchestration-engines/ai-agent/reasoning-action-loops/visual-reasoning/task-decomposition.md) — Breaks multi-step web tasks into sequential actions by reasoning over visual state. - [Visual Grounding Execution](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/agent-runtimes/agent-action-representations/action-coordinate-transformers/visual-grounding-execution.md) — Predicts click coordinates and scroll targets directly from pixel-level screen analysis without DOM reliance. - [Computer Use Agents](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-use-agents.md) — Controls a computer visually by predicting screen coordinates for clicking, typing, and scrolling. - [Compact Parameter Models](https://awesome-repositories.com/f/artificial-intelligence-ml/model-parameter-management/parameter-estimation-methods/transformer-parameter-calculators/compact-parameter-models.md) — Uses a compact 7-billion-parameter transformer that fits on consumer GPUs for low-latency inference. - [On-Device GUI Agents](https://awesome-repositories.com/f/artificial-intelligence-ml/on-device-models/on-device-gui-agents.md) — Runs a compact 7-billion-parameter model locally on consumer hardware for computer-use tasks with low latency. - [Visual Computer-Use Agents](https://awesome-repositories.com/f/artificial-intelligence-ml/self-hosted-ai-models/visual-computer-use-agents.md) — Ships a visual computer-use agent model deployable via vLLM, LM Studio, or Ollama for full local control. - [Live-Website Agent Benchmarks](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/frameworks/reinforcement-learning-environments/reinforcement-learning-performance-visualizers/agent-performance-evaluators/live-website-agent-benchmarks.md) — Includes a reproducible evaluation framework that runs agent benchmarks on 609 real, live web-browsing tasks. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) - [On-Device Deployments](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-deployment-and-serving/local-and-on-device-inference/edge-ai-model-deployment/generative-ai-models/edge-deployment-platforms/on-device-deployments.md) — Runs locally on consumer hardware with a compact 7-billion-parameter size for low latency and privacy. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) - [Local Model Inference Servers](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-deployment-and-serving/local-and-on-device-inference/local-model-inference-servers.md) — Supports self-hosted serving via vLLM, LM Studio, or Ollama for full inference control. ### User Interface & Experience - [Visual Computer Controllers](https://awesome-repositories.com/f/user-interface-experience/scroll-interaction-handlers/scroll-state-controllers/scroll-access-controllers/visual-computer-controllers.md) — Perceives webpages and performs scrolling, typing, and clicking on predicted coordinates without accessibility trees. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) ### DevOps & Infrastructure - [Managed Model Endpoints](https://awesome-repositories.com/f/devops-infrastructure/cloud-deployment-automation/azure-deployment-automators/azure-infrastructure-deployments/managed-model-endpoints.md) — Ships a managed endpoint deployment option on Azure Foundry for API-based model inference. - [Managed Agent Endpoints](https://awesome-repositories.com/f/devops-infrastructure/cloud-deployment/model-endpoint-deployment/managed-agent-endpoints.md) — Deploys a computer-use model via Azure Foundry endpoint without managing infrastructure or downloading weights. - [Managed AI Endpoints](https://awesome-repositories.com/f/devops-infrastructure/cloud-hosting-services/managed-ai-endpoints.md) — Deploys the model on Azure Foundry without downloading weights or managing GPU infrastructure. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) - [Self-Hosted Agent Inference](https://awesome-repositories.com/f/devops-infrastructure/self-hosted-installations/self-hosted-agent-inference.md) — Runs the model on a GPU machine using vLLM, LM Studio, or Ollama for full inference control. - [Self-Hosted Inference Servers](https://awesome-repositories.com/f/devops-infrastructure/self-hosted-server-platforms/self-hosted-inference-servers.md) — Runs the model on a GPU machine using vLLM, LM Studio, or Ollama for full inference control. ([source](https://cdn.jsdelivr.net/gh/microsoft/fara@main/README.md)) ### Software Engineering & Architecture - [Live Website Agent Benchmarks](https://awesome-repositories.com/f/software-engineering-architecture/function-execution-timing/live-execution-monitoring/live-benchmark-monitoring/live-website-agent-benchmarks.md) — Provides a reproducible evaluation framework running agent benchmarks on 609 real, live web-browsing tasks. ### Testing & Quality Assurance - [Agent Performance Benchmarks](https://awesome-repositories.com/f/testing-quality-assurance/agent-performance-benchmarks.md) — Runs a benchmark of 609 web-browsing tasks comparing agent performance across scoring rubrics. ([source](https://microsoft.github.io/fara/docs/webtailbench_v1_v2_diff.html))