Qwen3.5-122B-A10B is the mid-size Mixture-of-Experts model in Qwen's natively multimodal Qwen3.5 series: 122B total and 10B active parameters, text and image input, a 256K-token context extensible to 1M, and thinking on by default. This is Qwen's official FP8 checkpoint, released under Apache 2.0.
Qwen3.5-122B-A10B is a 48-layer hybrid model: each block of three Gated DeltaNet linear-attention layers is followed by one gated full-attention layer with 32 query heads and 2 key-value heads, and every layer routes through a 256-expert Mixture-of-Experts with 1,024-wide experts that activates 8 routed experts plus one shared expert, so only about 10B of its 122B parameters are used per token. Trained with early vision-language fusion, it accepts images alongside text, thinks by default before answering (thinking can be disabled per request), and is tuned for tool calling and agentic use. It supports 201 languages and dialects, a 262,144-token native context extensible to 1,010,000 tokens with YaRN, and includes a multi-token-prediction head for speculative decoding.
This is Qwen's official FP8 quantization of the post-trained model, fine-grained with a block size of 128, which Qwen reports as nearly identical in quality to the original weights. Qwen3.5-122B-A10B was released on February 24, 2026, alongside the 35B-A3B and 27B models, about a week after the flagship 397B-A17B, and its weights are released under the Apache 2.0 license. Full details are in Qwen's announcement and the model card.
A Dedicated Endpoint is your own deployment of Qwen3.5 122B A10B FP8, on an inference stack that Modal tunes and autoscales. Bring fine-tuned weights if you have them. Compute is billed by the second, and only while it runs.
modal endpoint create --model Qwen/Qwen3.5-122B-A10B-FP8