Qwen3.5

Name: Qwen3.5
Author: Alibaba

AlibabaOpen WeightsPending Human Review

Qwen3.5 is a native vision-language foundation-model family combining Gated DeltaNet and gated attention. The initial 397B-A17B MoE model was followed by 122B-A10B, 35B-A3B, and dense sizes from 27B to 0.8B. Early fusion of text and visual data supports images, video, coding, reasoning, and tool-based agents, with expanded coverage of 201 languages and dialects. Apache-2.0 weights permit local use. This page groups the open sizes; separately positioned Plus and Flash hosted tiers have different service limits and their own pages.

2026-02-16

397B total, 17B activated (initial flagship); 0.8B–122B additional sizes

Hybrid Gated DeltaNet/gated-attention Transformer; MoE and dense variants with vision encoder

Apache-2.0

Compare with other models

Specifications

Parameters: 397B total, 17B activated (initial flagship); 0.8B–122B additional sizes
Architecture: Hybrid Gated DeltaNet/gated-attention Transformer; MoE and dense variants with vision encoder
License: Apache-2.0
Context Window: 262,144 tokens
Type: multimodal
Modalities: textimagevideo

Benchmark Scores

Advanced Specifications

Model Family: Qwen
API Access: Not Available
Chat Interface: Not Available
Multilingual Support: Yes
Variants: Qwen3.5-397B-A17B (2026-02-16)Qwen3.5-122B-A10B (2026-02-24)Qwen3.5-35B-A3B (2026-02-24)Qwen3.5-27B (2026-02-24)Qwen3.5-9B (2026-03-02)Qwen3.5-4B (2026-03-02)Qwen3.5-2B (2026-03-02)Qwen3.5-0.8B (2026-03-02)Released Base checkpoints

Capabilities & Limitations

Capabilities: reasoningcode generationtool usevisual understandingmultilinguallong-context modeling
Known Limitations: Self-hosting requires sufficient memory for all model weights, including inactive experts.Reasoning and agent benchmark results depend on the harness and sampling settings.Parameter counts and memory requirements vary substantially across sizes.The native 262K context differs from hosted Plus/Flash limits and optional YaRN extension.
Notable Use Cases: coding assistantsresearch agentsdocument analysisvisual codinglocal multimodal assistants
Function Calling Support: Yes
Tool Use Support: Yes

Resources

Related Models

IndustryLLM

Alibaba

IndustryLLM is Alibaba’s open-weight language model for industrial procurement and engineering-standard reasoning. It adapts Qwen3.5-35B-A3B-Base through continued pre-training on industrial documents and transactions followed by task-focused supervised fine-tuning. The 35.95B-parameter hybrid mixture-of-experts network activates approximately 3B per token. Training targets informal buyer language, ambiguous product codes, contradictory specifications, and evidence-based constraint evaluation. Apache-licensed BF16 weights are published. Its 262K architectural context is distinct from 12,288-token packed fine-tuning sequences; this record uses the supported training length conservatively. The September 25 paper predates later discovery-index coverage.

Typetext

Parameters35.95B total, approximately 3B active

2026-09-25

Open Weights

Details Compare

Qwen-Audio-3.1-Realtime

Alibaba

Qwen-Audio-3.1-Realtime is a proprietary language-capable speech conversation model for fluid real-time dialogue. It combines speech reasoning with duplex interaction, allowing conversational responses rather than only transcription or speech synthesis. The 3.1 Plus update supports a 262,144-token context, function calling, web search, voice cloning, and additional system voices, while retaining the 3.0 Plus integration protocol. Hosted streaming APIs provide access; no downloadable weights or checkpoint parameter count is verified.

Qwen3.8-Omni-Flash

Alibaba

Qwen3.8-Omni-Flash is a native omnimodal agent model built on the Qwen3.8-Flash-Next architecture. It accepts text, images, audio, and video in a million-token context, extending coding, GUI operation, and knowledge work to audio-visual productivity. Standard APIs return text, while the subsequently available Realtime variant supports text and speech output, multichannel audio, video aggregation, and remote MCP tools. Hosted access and companion plugins support video editing, multimedia summarization, and real-time conversations; its weights are not publicly released.