证据：Qwen documents 119 languages and dialects for Qwen3. For Qwen3-2507, the maximum assistant message length is 81,920 tokens for thinking models and 16,384 tokens for instruct models; Qwen3-2507 supports 256K context and can be extended to 1M tokens.
条件：Re-verified on 2026-10-20 against the current model card: 32,768 tokens natively; 131,072 tokens only when YaRN RoPE scaling is enabled by the serving framework, so the native limit is recorded here.
证据：Front matter: license: apache-2.0. Features list: Number of Parameters: 235B in total and 22B activated; Context Length: 32,768 natively and 131,072 tokens with YaRN. Long-text section: Qwen3 natively supports context lengths of up to 32,768 tokens, and the model's performance was validated up to 131,072 tokens using the YaRN method. The Hugging Face model API reports the repository as public, not gated and not disabled, and the repository exposes safetensors weights.
证据：Qwen documents 119 languages and dialects for Qwen3. For Qwen3-2507, the maximum assistant message length is 81,920 tokens for thinking models and 16,384 tokens for instruct models; Qwen3-2507 supports 256K context and can be extended to 1M tokens.
证据：Qwen states that Qwen3 was released on April 29, 2025, open-weights two MoE models and six dense models under Apache 2.0. The model tables list layers, attention heads, experts, and 128K context for the release family.
lifecycle_status
公开事实
active
高置信度，已绑定来源与证据。
条件：Set on 2026-10-20 because the Qwen3-235B-A22B weights remain publicly downloadable from the official Hugging Face repository, which the API reports as neither gated nor disabled; Qwen does not publish a formal lifecycle table, so this reflects current availability rather than a vendor-stated status.
证据：Front matter: license: apache-2.0. Features list: Number of Parameters: 235B in total and 22B activated; Context Length: 32,768 natively and 131,072 tokens with YaRN. Long-text section: Qwen3 natively supports context lengths of up to 32,768 tokens, and the model's performance was validated up to 131,072 tokens using the YaRN method. The Hugging Face model API reports the repository as public, not gated and not disabled, and the repository exposes safetensors weights.
证据：Front matter: license: apache-2.0. Features list: Number of Parameters: 235B in total and 22B activated; Context Length: 32,768 natively and 131,072 tokens with YaRN. Long-text section: Qwen3 natively supports context lengths of up to 32,768 tokens, and the model's performance was validated up to 131,072 tokens using the YaRN method. The Hugging Face model API reports the repository as public, not gated and not disabled, and the repository exposes safetensors weights.
证据：Qwen states that Qwen3 uses approximately 36 trillion pre-training tokens across 119 languages and dialects. The pre-training process has three stages; post-training has four stages: long chain-of-thought cold start, reasoning-based reinforcement learning, thinking mode fusion, and general reinforcement learning.
证据：Qwen states that Qwen3 uses approximately 36 trillion pre-training tokens across 119 languages and dialects. The pre-training process has three stages; post-training has four stages: long chain-of-thought cold start, reasoning-based reinforcement learning, thinking mode fusion, and general reinforcement learning.
证据：Qwen states that Qwen3 was released on April 29, 2025, open-weights two MoE models and six dense models under Apache 2.0. The model tables list layers, attention heads, experts, and 128K context for the release family.
证据：Qwen states that Qwen3 was released on April 29, 2025, open-weights two MoE models and six dense models under Apache 2.0. The model tables list layers, attention heads, experts, and 128K context for the release family.
证据：Qwen documents 119 languages and dialects for Qwen3. For Qwen3-2507, the maximum assistant message length is 81,920 tokens for thinking models and 16,384 tokens for instruct models; Qwen3-2507 supports 256K context and can be extended to 1M tokens.
证据：Front matter: license: apache-2.0. Features list: Number of Parameters: 235B in total and 22B activated; Context Length: 32,768 natively and 131,072 tokens with YaRN. Long-text section: Qwen3 natively supports context lengths of up to 32,768 tokens, and the model's performance was validated up to 131,072 tokens using the YaRN method. The Hugging Face model API reports the repository as public, not gated and not disabled, and the repository exposes safetensors weights.
证据：Qwen states that Qwen3 uses approximately 36 trillion pre-training tokens across 119 languages and dialects. The pre-training process has three stages; post-training has four stages: long chain-of-thought cold start, reasoning-based reinforcement learning, thinking mode fusion, and general reinforcement learning.
Three-stage pretraining: S1 over 30T tokens at 4K context; S2 adds 5T knowledge-intensive data; S3 uses high-quality long-context data to extend context to 32K.
证据：Qwen states that Qwen3 uses approximately 36 trillion pre-training tokens across 119 languages and dialects. The pre-training process has three stages; post-training has four stages: long chain-of-thought cold start, reasoning-based reinforcement learning, thinking mode fusion, and general reinforcement learning.
条件：Re-verified on 2026-10-20; the Hugging Face repository for Qwen3-235B-A22B is public, not gated and exposes safetensors weights, so the open weights remain downloadable.
证据：Front matter: license: apache-2.0. Features list: Number of Parameters: 235B in total and 22B activated; Context Length: 32,768 natively and 131,072 tokens with YaRN. Long-text section: Qwen3 natively supports context lengths of up to 32,768 tokens, and the model's performance was validated up to 131,072 tokens using the YaRN method. The Hugging Face model API reports the repository as public, not gated and not disabled, and the repository exposes safetensors weights.