{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"baseten","slug":"baseten","name":"Baseten","type":"platform","url":"https://baseten.co","page_url":"https://unfragile.ai/baseten","categories":["deployment-infra"],"tags":[],"pricing":{"model":"usage","free":false,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"baseten__cap_0","uri":"capability://automation.workflow.gpu.accelerated.model.inference.with.per.minute.billing","name":"gpu-accelerated model inference with per-minute billing","description":"Deploys models on dedicated GPU instances (T4, L4, A10G, A100, H100, B200) with granular per-minute billing down to the minute. Infrastructure automatically provisions and tears down compute resources based on deployment lifecycle, with pricing ranging from $0.01/min for T4 to $0.17/min for B200. Supports both single-model and multi-GPU configurations with transparent pricing visibility per hardware tier.","intents":["Deploy a fine-tuned LLM on A100 GPU and pay only for active inference time","Run batch image generation jobs on H100s without long-term infrastructure contracts","Test model performance across different GPU tiers to find cost-optimal hardware","Scale inference workloads from development (T4) to production (A100/H100) with predictable per-minute costs"],"best_for":["ML teams building production inference APIs with variable traffic patterns","Startups avoiding upfront GPU infrastructure investment","Researchers comparing model performance across hardware tiers"],"limitations":["Per-minute granularity means short-lived inferences (< 1 min) are billed as full minute","No spot/preemptible instance pricing available — only on-demand rates","Egress bandwidth pricing not documented; potential hidden costs for large output transfers","Cold start latency claimed as 'blazing-fast' but no specific SLA or latency guarantees published"],"requires":["Active Baseten account with payment method","Model packaged in Truss format or compatible container","API key for authentication"],"input_types":["model weights (PyTorch, TensorFlow, ONNX)","inference requests (JSON via REST API)"],"output_types":["inference results (JSON)","usage metrics (per-minute billing records)"],"categories":["automation-workflow","infrastructure-as-code"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_1","uri":"capability://automation.workflow.cpu.based.inference.with.6.instance.tiers","name":"cpu-based inference with 6 instance tiers","description":"Provisions CPU-only instances ranging from 1vCPU/2GB RAM ($0.00058/min) to 16vCPU/64GB RAM ($0.01382/min) for models that don't require GPU acceleration. Uses standard cloud compute instances with per-minute billing, enabling cost-effective serving of lightweight models, embeddings, or CPU-optimized inference workloads without GPU overhead.","intents":["Deploy lightweight embedding models or text classification on CPU without GPU cost","Run inference for models optimized for CPU (e.g., ONNX quantized models)","Scale CPU-based APIs with predictable per-minute costs","Test model performance on CPU before GPU deployment"],"best_for":["Teams serving lightweight NLP models (embeddings, classifiers, tokenizers)","Cost-conscious deployments where GPU acceleration isn't needed","Development and testing environments"],"limitations":["CPU inference significantly slower than GPU for large models (LLMs, diffusion models)","No CPU-specific optimization details provided (no mention of SIMD, quantization support, or batching strategies)","Maximum 16vCPU instance may be insufficient for high-concurrency CPU-bound workloads","No reserved capacity or spot pricing — only on-demand per-minute billing"],"requires":["Model compatible with CPU execution (no CUDA dependencies)","Baseten account with payment method"],"input_types":["model weights (PyTorch, TensorFlow, ONNX)","inference requests (JSON)"],"output_types":["inference results (JSON)","usage metrics"],"categories":["automation-workflow","infrastructure-as-code"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_10","uri":"capability://tool.use.integration.multi.provider.model.api.access.with.unified.interface","name":"multi-provider model api access with unified interface","description":"Aggregates multiple LLM providers (DeepSeek, Kimi, NVIDIA Nemotron, GLM) under a single Baseten API interface, enabling developers to switch between models without changing application code. Provides unified authentication, request/response formatting, and error handling across providers. Simplifies provider evaluation and migration by standardizing API contracts.","intents":["Evaluate multiple LLM providers (DeepSeek, Kimi, NVIDIA) without integrating each separately","Switch between LLM providers without changing application code","Compare model performance and cost across providers using identical API","Reduce vendor lock-in by abstracting provider-specific APIs"],"best_for":["Developers building LLM applications who want provider flexibility","Teams evaluating multiple model providers before committing","Applications requiring provider switching for cost optimization or availability"],"limitations":["Unified API specification not documented — no details on request/response format or error handling","Model selection mechanism unclear — no documentation on how to specify provider or model version","Provider coverage limited compared to OpenAI/Anthropic — missing GPT-4, Claude, Gemini","No information on provider-specific features, rate limits, or SLAs","Unclear if unified interface supports all provider-specific parameters (e.g., vision, function calling)"],"requires":["Baseten account with API access","API key for authentication","Understanding of unified API contract (unknown)"],"input_types":["text prompts (JSON)","model selection parameter (format unknown)"],"output_types":["text completions (JSON)"],"categories":["tool-use-integration","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_11","uri":"capability://safety.moderation.compliance.and.security.certifications.soc.2.hipaa","name":"compliance and security certifications (soc 2, hipaa)","description":"Provides SOC 2 Type II and HIPAA compliance certifications across all tiers (Basic and above), enabling deployment of healthcare and regulated workloads. Enterprise tier adds advanced security features including custom RBAC with Teams, enhanced data protection, and compliance controls. Certifications enable organizations to meet regulatory requirements without additional security infrastructure.","intents":["Deploy healthcare AI models (e.g., medical imaging, clinical NLP) with HIPAA compliance","Meet SOC 2 audit requirements for enterprise customers","Implement role-based access control for multi-team organizations","Satisfy regulatory compliance requirements without building custom security infrastructure"],"best_for":["Healthcare and regulated industry teams requiring HIPAA compliance","Enterprise organizations with SOC 2 audit requirements","Multi-team organizations requiring fine-grained access control"],"limitations":["Compliance scope not documented — unclear which Baseten services are HIPAA-compliant (training, inference, storage)","HIPAA implementation details unknown — no information on encryption, audit logging, or data handling","Advanced security features (Enterprise tier) not specified — unclear what additional controls are provided","Custom RBAC details unknown — no documentation on role definitions, permission model, or team management","No mention of compliance monitoring, audit trails, or compliance reporting tools"],"requires":["Basic tier or higher for SOC 2/HIPAA","Enterprise tier for advanced security and custom RBAC","Compliance agreement with Baseten"],"input_types":["model deployment requests","user access requests"],"output_types":["compliance certifications","audit logs (format unknown)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_12","uri":"capability://planning.reasoning.forward.deployed.engineering.support.for.production.optimization","name":"forward-deployed engineering support for production optimization","description":"Provides hands-on engineering support from Baseten's team for production optimization, model tuning, and deployment best practices. Available on Pro and Enterprise tiers, enabling organizations to leverage Baseten expertise for rapid prototyping and production hardening. Support includes model optimization, performance tuning, and architecture guidance.","intents":["Get expert guidance on optimizing model performance for production workloads","Accelerate time-to-production through hands-on engineering support","Implement best practices for inference optimization, caching, and scaling","Troubleshoot production issues with direct access to Baseten engineers"],"best_for":["Teams building production ML systems who benefit from expert guidance","Organizations with complex inference requirements requiring optimization","Startups accelerating time-to-market with expert support"],"limitations":["Support scope and SLA not documented — unclear what services are included or response times","Availability limited to Pro and Enterprise tiers — not available for Basic tier","Support process and engagement model unknown — no documentation on how to request support","No information on support hours, availability, or escalation procedures","Unclear if support covers custom models, Model APIs, or both"],"requires":["Pro tier or higher subscription","Baseten account with support access"],"input_types":["support requests (format unknown)","model and deployment details"],"output_types":["optimization recommendations","implementation guidance"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_13","uri":"capability://safety.moderation.99.99.uptime.sla.with.global.capacity","name":"99.99% uptime sla with global capacity","description":"Guarantees 99.99% uptime for deployed inference endpoints across all tiers (Basic and above), with global capacity distribution enabling low-latency serving across regions. Infrastructure is designed for high availability with automatic failover and redundancy. Enterprise tier enables custom global regions and full data residency control for compliance-sensitive workloads.","intents":["Deploy production inference APIs with 99.99% uptime guarantee","Serve users globally with low-latency inference from distributed capacity","Ensure inference availability during infrastructure failures with automatic failover","Meet SLA requirements for mission-critical applications"],"best_for":["Production applications requiring high availability and uptime guarantees","Global applications requiring low-latency inference across regions","Mission-critical workloads where downtime has significant business impact"],"limitations":["SLA details not documented — no information on uptime measurement, credits, or exclusions","Global capacity distribution not specified — no details on available regions or latency guarantees","Failover mechanism unknown — no documentation on automatic failover, recovery time, or data consistency","Custom global regions (Enterprise tier) not detailed — no list of available regions or setup process","No mention of disaster recovery, backup, or business continuity features"],"requires":["Basic tier or higher for 99.99% uptime SLA","Enterprise tier for custom global regions and data residency"],"input_types":["inference requests"],"output_types":["inference results","uptime metrics"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_2","uri":"capability://tool.use.integration.model.api.marketplace.with.pre.optimized.inference.endpoints","name":"model api marketplace with pre-optimized inference endpoints","description":"Hosts a curated library of pre-optimized model APIs (DeepSeek V4, Kimi K2.6, NVIDIA Nemotron, GLM 5, Whisper Large V3, ComfyUI workflows) available for instant testing and production use with per-token pricing. Models are pre-deployed and optimized with custom kernels and advanced decoding techniques, eliminating deployment complexity. Pricing varies by model (e.g., DeepSeek V4: $1.74/1M input tokens, $3.48/1M output tokens) with KV cache optimization for cached input tokens ($0.145/1M).","intents":["Use a production-ready LLM API without deploying or managing infrastructure","Test multiple LLM providers (DeepSeek, Kimi, NVIDIA, GLM) from a single platform","Leverage KV cache pricing to reduce costs for repeated context (e.g., document Q&A)","Access pre-optimized image generation (ComfyUI) and audio models (Whisper) without custom deployment"],"best_for":["Developers building LLM applications who want managed inference without deployment overhead","Teams evaluating multiple model providers with unified API","Applications with high token volume that benefit from KV cache cost optimization"],"limitations":["Limited model selection compared to OpenAI/Anthropic (no GPT-4, Claude, or Gemini)","Per-token pricing model differs from per-minute GPU billing — cost unpredictable for variable-length outputs","No information on rate limits, concurrency limits, or request timeout policies","Model library details unknown — no documentation on available models, versions, or update frequency","No mention of fine-tuning or custom model deployment via Model API tier"],"requires":["Baseten account with payment method","API key for authentication","Understanding of per-token pricing model"],"input_types":["text prompts (JSON)","image generation parameters (JSON for ComfyUI)","audio files (Whisper)"],"output_types":["text completions (JSON)","generated images","transcriptions (JSON)"],"categories":["tool-use-integration","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_3","uri":"capability://automation.workflow.truss.model.packaging.and.containerization","name":"truss model packaging and containerization","description":"Open-source model packaging framework that standardizes model deployment across Baseten and other platforms. Truss wraps models with dependencies, inference logic, and configuration in a portable container format, enabling one-command deployment to Baseten infrastructure. Abstracts away Docker/Kubernetes complexity while maintaining full control over model serving code, dependencies, and resource requirements.","intents":["Package a custom fine-tuned model with dependencies and deploy to Baseten in one command","Version control model code, weights, and inference logic together","Test model locally before deploying to Baseten production","Migrate models between Baseten and other inference platforms using standard Truss format"],"best_for":["ML engineers building custom models and inference pipelines","Teams wanting to avoid vendor lock-in through open-source packaging","Developers preferring infrastructure-as-code approach to model deployment"],"limitations":["Truss repository and documentation details unknown — no link to GitHub or API reference provided","No information on supported frameworks (PyTorch, TensorFlow, JAX, etc.) or Python versions","Portability claims unclear — no documentation on exporting Truss models to other platforms","Local testing capability mentioned but testing workflow and debugging tools not documented","No mention of model versioning, rollback, or A/B testing within Truss framework"],"requires":["Python 3.x (version unknown)","Model weights and inference code","Baseten CLI or API for deployment"],"input_types":["model weights (PyTorch, TensorFlow, etc.)","Python inference code","requirements.txt or dependency manifest"],"output_types":["Truss container image","deployed inference endpoint"],"categories":["automation-workflow","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_4","uri":"capability://automation.workflow.auto.scaling.inference.with.unlimited.concurrency.pro.tier","name":"auto-scaling inference with unlimited concurrency (pro tier)","description":"Automatically scales inference endpoints based on request volume, provisioning additional GPU/CPU instances to handle traffic spikes without manual intervention. Pro tier enables 'unlimited autoscaling' with no documented concurrency limits or scaling policies. Scaling mechanism abstracts infrastructure management, allowing developers to focus on model optimization rather than capacity planning.","intents":["Deploy an inference API that handles traffic spikes without manual scaling","Avoid over-provisioning by scaling down during low-traffic periods","Support variable-traffic applications (e.g., user-triggered image generation) without capacity planning","Achieve high availability without managing load balancers or orchestration"],"best_for":["Applications with unpredictable or bursty traffic patterns","Teams without DevOps expertise to manage Kubernetes or load balancing","Production workloads requiring high availability and elasticity"],"limitations":["Scaling mechanism completely undocumented — no details on scaling policies, metrics, or decision logic","No published concurrency limits, queue management strategy, or request timeout policies","Cold start latency during scale-up events not quantified ('blazing-fast' claimed but no SLA)","Scaling costs unclear — no documentation on how auto-scaling affects per-minute billing","No control over scaling parameters (min/max replicas, scale-up/down thresholds, cooldown periods)","Unlimited autoscaling claim may have undocumented limits or require Enterprise tier for true unlimited scaling"],"requires":["Pro tier subscription or higher","Model deployed on Baseten infrastructure","API key for monitoring and management"],"input_types":["inference requests (JSON)"],"output_types":["inference results (JSON)","scaling metrics (unknown format)"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_5","uri":"capability://automation.workflow.model.versioning.and.production.deployment.management","name":"model versioning and production deployment management","description":"Manages multiple versions of deployed models with production-ready versioning controls, enabling safe rollouts, rollbacks, and A/B testing. Supports deploying different model versions simultaneously and routing traffic between them. Integrates with monitoring to track performance per version, facilitating gradual rollouts and quick rollback on degradation.","intents":["Deploy a new model version to production while keeping the previous version active for rollback","Run A/B tests comparing two model versions with traffic splitting","Gradually roll out a new model version (canary deployment) to detect issues early","Track inference metrics per model version to identify performance regressions"],"best_for":["Production ML teams managing model lifecycle and continuous improvement","Teams requiring safe deployment practices with rollback capability","Applications where model performance directly impacts user experience"],"limitations":["Versioning details completely undocumented — no information on version naming, storage, or retention policies","No documentation on traffic splitting, canary deployment, or A/B testing mechanics","Rollback procedures and rollback time not specified","No mention of version-specific monitoring, metrics, or performance comparison tools","Unclear if versioning applies to both custom models and Model API endpoints"],"requires":["Model deployed on Baseten infrastructure","Understanding of versioning workflow (unknown)"],"input_types":["model weights and code (Truss format or container)"],"output_types":["versioned inference endpoints","deployment status and metrics"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_6","uri":"capability://safety.moderation.monitoring.and.observability.for.deployed.models","name":"monitoring and observability for deployed models","description":"Provides built-in monitoring and observability for inference endpoints, tracking performance metrics, latency, error rates, and usage patterns. Integrates with deployment versioning to enable per-version performance comparison. Monitoring is included in all tiers (Basic and above) with advanced observability features available in Enterprise tier.","intents":["Monitor inference latency and error rates in production to detect performance degradation","Compare performance metrics across model versions to validate improvements","Track API usage and costs per endpoint for billing and optimization","Set up alerts for SLA violations or anomalous inference patterns"],"best_for":["Production ML teams requiring visibility into model performance","Teams managing multiple model versions and needing performance comparison","Cost-conscious teams tracking inference costs and optimizing resource usage"],"limitations":["Monitoring capabilities completely undocumented — no details on metrics, dashboards, or alerting","No information on metric retention, sampling rates, or data export options","Advanced observability features (Enterprise tier) not specified","No mention of custom metrics, distributed tracing, or integration with external monitoring tools","Unclear if monitoring covers both custom deployments and Model API endpoints"],"requires":["Model deployed on Baseten infrastructure","Baseten account with monitoring access"],"input_types":["inference requests and responses (automatic collection)"],"output_types":["performance metrics (format unknown)","dashboards (format unknown)","alerts (format unknown)"],"categories":["safety-moderation","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_7","uri":"capability://automation.workflow.one.click.training.to.inference.deployment.pipeline","name":"one-click training-to-inference deployment pipeline","description":"Integrates training and inference workflows, enabling models trained on Baseten to be deployed as inference endpoints with a single click. Eliminates manual model export, packaging, and deployment steps by maintaining model continuity from training to production. Supports deploying trained models directly to GPU inference infrastructure without intermediate steps.","intents":["Train a custom model and immediately deploy it to production inference without manual packaging","Iterate on model training and quickly test inference performance on production hardware","Reduce time-to-production for fine-tuned models from hours to minutes","Maintain model lineage from training to inference for reproducibility"],"best_for":["ML teams building and iterating on custom models","Rapid prototyping and experimentation workflows","Teams without dedicated MLOps infrastructure"],"limitations":["Training capabilities and supported frameworks not documented","Deployment mechanism from training to inference not specified","No information on training cost, duration, or hardware options","Unclear if training supports distributed training, hyperparameter tuning, or experiment tracking","No mention of training data management, versioning, or reproducibility features"],"requires":["Training data and model code compatible with Baseten training","Baseten account with training and inference access"],"input_types":["training data (format unknown)","model code (framework unknown)"],"output_types":["trained model weights","deployed inference endpoint"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_8","uri":"capability://automation.workflow.self.hosted.and.hybrid.deployment.options","name":"self-hosted and hybrid deployment options","description":"Supports deploying models on customer-controlled infrastructure (self-hosted) or hybrid configurations combining self-hosted deployments with on-demand flex capacity on Baseten Cloud. Enables data residency control, compliance requirements, and reduced vendor lock-in. Enterprise tier includes full self-hosted support with custom global regions and data residency guarantees.","intents":["Deploy models on-premises or in customer VPC for data privacy and compliance","Maintain control over model weights and inference data without cloud exposure","Burst to Baseten Cloud capacity during traffic spikes while keeping baseline on self-hosted infrastructure","Meet data residency requirements (HIPAA, GDPR) by hosting in specific regions"],"best_for":["Enterprise teams with strict data residency or compliance requirements","Organizations requiring on-premises model deployment for security","Hybrid cloud strategies combining self-hosted and cloud capacity"],"limitations":["Self-hosted deployment details completely undocumented — no information on supported infrastructure, setup, or management","Hybrid deployment mechanism not specified — unclear how traffic routing and failover work","Custom global regions (Enterprise tier) not detailed — no list of available regions or setup process","No documentation on self-hosted monitoring, scaling, or support","Self-hosted option limited to Enterprise tier — not available for Basic or Pro tiers"],"requires":["Enterprise tier subscription","Customer infrastructure (on-premises or VPC)","Network connectivity between self-hosted and Baseten Cloud (for hybrid)","Deployment and infrastructure management expertise"],"input_types":["model weights and code","infrastructure configuration (format unknown)"],"output_types":["self-hosted inference endpoint","hybrid deployment configuration"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__cap_9","uri":"capability://image.visual.comfyui.workflow.deployment.for.image.generation","name":"comfyui workflow deployment for image generation","description":"Supports deploying ComfyUI workflows as production inference endpoints, enabling complex image generation pipelines with multiple model stages (e.g., LoRA loading, upscaling, inpainting). Abstracts ComfyUI complexity by packaging workflows as Baseten endpoints with standard API interface. Enables non-technical users to deploy sophisticated image generation without writing custom inference code.","intents":["Deploy a ComfyUI workflow (e.g., Stable Diffusion + LoRA + upscaling) as a production API","Run image generation pipelines without writing custom Python inference code","Scale ComfyUI workflows to handle high-volume image generation requests","Version and manage multiple ComfyUI workflows as separate endpoints"],"best_for":["Teams building image generation applications using ComfyUI","Non-technical users deploying pre-built ComfyUI workflows","Applications requiring complex multi-stage image generation pipelines"],"limitations":["ComfyUI support details completely undocumented — no information on workflow format, supported nodes, or limitations","No documentation on workflow versioning, parameter customization, or API interface","Unclear if ComfyUI workflows can be combined with other model types or custom inference code","No mention of performance optimization, batching, or caching for image generation","GPU requirements for ComfyUI workflows not specified"],"requires":["ComfyUI workflow (JSON or compatible format)","Model weights for workflow nodes (Stable Diffusion, LoRA, etc.)","Baseten account with image generation capability"],"input_types":["image generation parameters (JSON)","input images (for inpainting, upscaling)"],"output_types":["generated images (PNG, JPEG, etc.)"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"baseten__headline","uri":"capability://deployment.infra.ai.model.deployment.platform","name":"ai model deployment platform","description":"Baseten is a robust platform for deploying AI models as auto-scaling API endpoints, featuring GPU support and optimized inference engines for production use.","intents":["best AI model deployment platform","AI model deployment for scalable applications","production-ready AI model hosting","deploying machine learning models as APIs","auto-scaling AI inference services"],"best_for":["scalable AI applications","production environments"],"limitations":[],"requires":[],"input_types":["AI models"],"output_types":["API endpoints"],"categories":["deployment-infra"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":56,"verified":false,"data_access_risk":"high","permissions":["Active Baseten account with payment method","Model packaged in Truss format or compatible container","API key for authentication","Model compatible with CPU execution (no CUDA dependencies)","Baseten account with payment method","Baseten account with API access","Understanding of unified API contract (unknown)","Basic tier or higher for SOC 2/HIPAA","Enterprise tier for advanced security and custom RBAC","Compliance agreement with Baseten"],"failure_modes":["Per-minute granularity means short-lived inferences (< 1 min) are billed as full minute","No spot/preemptible instance pricing available — only on-demand rates","Egress bandwidth pricing not documented; potential hidden costs for large output transfers","Cold start latency claimed as 'blazing-fast' but no specific SLA or latency guarantees published","CPU inference significantly slower than GPU for large models (LLMs, diffusion models)","No CPU-specific optimization details provided (no mention of SIMD, quantization support, or batching strategies)","Maximum 16vCPU instance may be insufficient for high-concurrency CPU-bound workloads","No reserved capacity or spot pricing — only on-demand per-minute billing","Unified API specification not documented — no details on request/response format or error handling","Model selection mechanism unclear — no documentation on how to specify provider or model version","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.7,"quality":0.9,"ecosystem":0.15000000000000002,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.3,"quality":0.25,"ecosystem":0.15,"match_graph":0.25,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.013Z","last_scraped_at":null,"last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=baseten","compare_url":"https://unfragile.ai/compare?artifact=baseten"}},"signature":"Yg49jrtLyx5A/eEgJEwJ2fo8jnOIngEXTN5+ii1To+kPuzKXwxh31PxvRpaGJpojyXbUpq/HMqEYw/rp/N5AAA==","signedAt":"2026-06-20T19:56:25.036Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/baseten","artifact":"https://unfragile.ai/baseten","verify":"https://unfragile.ai/api/v1/verify?slug=baseten","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}