FW-Qwen3.5-35B-A3B

Models available for use with Fireworks on Foundry deliver optimized, best-in-class performance on the Fireworks Inference Cloud. Fireworks on Foundry is a Non-Microsoft Product. The following terms apply to a Customer's use of Fireworks on Foundry: When you use Fireworks on Foundry, data is shared between Microsoft and Fireworks AI, Customer Data will be sent outside of Microsoft systems, Customer Data will not be processed pursuant to any Foundry data residency documentation, and different compliance and data handling rules will apply. See Trust Center - Fireworks AI for details. Customers are responsible for evaluating whether data sharing between Microsoft and Fireworks is appropriate for their organization's compliance requirements.

About this model

Qwen3.5-35B-A3B is a post-trained causal language model with a vision encoder that uses a hybrid stack of Gated DeltaNet and Gated Attention blocks with sparse Mixture-of-Experts routing, totaling 35B parameters with 3B activated per token. Key innovations include early-fusion multimodal training, scalable reinforcement learning across million-agent environments, and training infrastructure designed for near-100% multimodal efficiency. The model accepts text and image inputs and generates text, with support for 201 languages and dialects and strong performance on reasoning, coding, agents, and visual understanding. It supports a native 262,144-token context window and can be extended to 1,010,000 tokens.

Key model capabilities

• Unified vision-language foundation with early-fusion multimodal training
• Hybrid Gated DeltaNet and sparse Mixture-of-Experts architecture
• High-throughput inference with minimal latency and cost overhead
• Support for 201 languages and dialects
• 262,144-token native context, extensible to 1,010,000 tokens