Fireworks adds video and audio inputs for Qwen3 Omni and Molmo2, and AWS S3 BYOB training data

Video & Audio Input Models

You can now query multimodal models with video and audio inputs for video captioning, scene analysis, and multimodal question answering. Deploy models like Qwen3 Omni and Molmo2 to process video and audio content directly using the Chat Completions API.

See the Video & Audio Inputs guide for deployment instructions and code examples.

AWS S3 Integration for Training Datasets

Training datasets can now be stored in your own AWS S3 buckets using GCP-to-AWS OIDC federation. This Bring Your Own Bucket (BYOB) approach keeps your data private while enabling secure access during Supervised Fine-Tuning and Reinforcement Fine-Tuning jobs—no long-lived credentials required.

See the Secure Training (BYOB) documentation for IAM role setup and usage examples.

Just-In-Time (JIT) User Provisioning for SSO (Enterprise)

JIT user provisioning automatically creates user accounts when users sign in through SSO for the first time. Enable this when configuring your identity provider to eliminate manual user creation.

See the SSO documentation for setup instructions.

📚 Documentation Updates

• Video & Audio Inputs: New guide for processing video and audio with Qwen3 Omni and Molmo2 models (Video & Audio Inputs)
• AWS S3 Bucket Integration: BYOB dataset storage for training via OIDC federation (Secure Training)
• Rate Limits Clarification: Expanded documentation on adaptive serverless limits, upper bounds, and monitoring (Serverless rate limits)
• Anthropic-Compatible Thinking Parameter: Control reasoning with thinking parameter alongside reasoning_effort (Reasoning)
• Scaling from Zero Behavior: Deployments scaled to zero return 503 immediately with retry guidance (Autoscaling)
• Custom Model Configuration: Generation defaults via fireworks.json for LoRA adapters (Uploading Custom Models)
• Service Account Roles: Role assignment and management for service accounts (Service Accounts)
• Free RFT: Documentation highlighting free Reinforcement Fine-Tuning for models under 16B parameters
• Serverless Model Retrieval: Python SDK and curl examples for programmatically listing serverless models (FAQ)
• Function Tools in Responses API: Client-executed function tools following OpenAI-compatible format (Responses API)
• Autoscaling Metric: New prompt_tokens_per_second load target for prefill-heavy workloads (Autoscaling)
• firectl CLI Syntax: All documentation updated to firectl <noun> <verb> syntax

[accordion] Bug Fixes & Minor Improvements

• Model Library Enhancements: Skeleton loader for faster perceived loading, new filter categories for Embeddings, Reranks, and Agents (Web App)
• Service Account Roles: Role column and role assignment in the Users management page (Web App)
• JIT User Provisioning Toggle: Enable JIT provisioning when configuring SSO identity providers (Web App, Enterprise)
• Autoscaling Configuration: Autoscaling metric and threshold settings in deployment creation form (Web App)
• RFT Chunk Size: Configurable chunk size setting with tooltips for Reinforcement Fine-Tuning creation (Web App)
• Batch Inference Form: Redesigned step-based form flow for batch inference job creation (Web App)
• Full Precision Shape: "Full Precision" deployment shape option displayed in deployment configuration (Web App)
• Batch Inference Error Banner: Error banners now display on batch inference job detail pages (Web App)
• Audit Log Resource Column: Improved readability with tooltip for long resource names (Web App)
• API Key Error Messages: Clearer permission error messages when users lack API key management access (Web App)
• Fine-Tuning Cost Display: Fixed estimated cost display for fine-tuning jobs (Web App)
• Batch Inference Max Tokens: Fallback max value increased to 128,000 tokens (Web App)
• Playground Model Select: Improved performance for model selection dropdown (Web App)
• Embedding Model Shapes: Deployment shapes correctly filtered by embedding model type (Web App)
• Audit Logs Command: New firectl audit-logs list command for querying audit logs (CLI)