Model comparison
Qwen3-Next-80B-A3B-Thinking vs o3
Capability radar, value bars, and side-by-side posture.
Qwen3-Next-80B-A3B-Thinking
Not scored
value score
o3
93
value score
Qwen3-Next-80B-A3B-Thinking ctx
—
o3 ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- o393
Attribute tape
| Field | Qwen3-Next-80B-A3B-Thinking | o3 |
|---|---|---|
| Developer | Qwen | OpenAI |
| Context | — | 200K |
| Modalities | Text, Code | |
| Openness | — | Proprietary Model |
| Speed | — | — |
| Price | — | $10/1M in · $40/1M out |
| Value score | — | 93 |
| Summary | Qwen3-Next-80B-A3B-Thinking is a reasoning-focused large language model from Alibaba's Qwen team that always operates in thinking mode, generating a step-by-step reasoning trace before producing its final answer. It uses a hybrid Gated DeltaNet and Gated Attention architecture with 80B total parameters but only 3B active per token, and supports a native 262K-token context window extensible to approximately 1M tokens. | OpenAI reasoning model focused on hard problems. |