Model Comparison

Qwen3.5-397B-A17B vs Qwen3-Next-80B-A3B-ThinkingWhich is better in 2026?

Qwen3.5-397B-A17B significantly outperforms across most benchmarks. Qwen3-Next-80B-A3B-Thinking is 2.8x cheaper per token.

Verdict: Qwen3.5-397B-A17B vs Qwen3-Next-80B-A3B-Thinking — which is better?

Qwen3.5-397B-A17B (by Alibaba Cloud / Qwen Team) and Qwen3-Next-80B-A3B-Thinking (by Alibaba Cloud / Qwen Team) are two of the AI models people compare most. Here is how they stack up on benchmarks, price and capabilities, and which one to pick in 2026.

Qwen3.5-397B-A17B outperforms in 10 benchmarks (GPQA, HMMT25, IFEval, Include, LiveCodeBench v6, MMLU-Pro, MMLU-ProX, MMLU-Redux, PolyMATH, SuperGPQA), while Qwen3-Next-80B-A3B-Thinking is better at 0 benchmarks. Qwen3.5-397B-A17B significantly outperforms across most benchmarks.

On price, Qwen3-Next-80B-A3B-Thinking is roughly 2.8x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Qwen3.5-397B-A17B also accepts a larger context window (262,144 input tokens), making it the stronger choice for long documents and large codebases.

Choose Qwen3.5-397B-A17B if…

  • you want the strongest raw capability — it leads on 10 of 10 shared benchmarks
  • you process long inputs — it offers a 262,144 token context window
  • you want the most recent training data — it shipped Feb 2026

Choose Qwen3-Next-80B-A3B-Thinking if…

  • cost matters — it's about 2.8x cheaper per token
Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

10 benchmarks

Qwen3.5-397B-A17B outperforms in 10 benchmarks (GPQA, HMMT25, IFEval, Include, LiveCodeBench v6, MMLU-Pro, MMLU-ProX, MMLU-Redux, PolyMATH, SuperGPQA), while Qwen3-Next-80B-A3B-Thinking is better at 0 benchmarks.

Qwen3.5-397B-A17B significantly outperforms across most benchmarks.

Mon Jun 22 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Qwen3-Next-80B-A3B-Thinking costs less

For input processing, Qwen3.5-397B-A17B ($0.60/1M tokens) is 4.0x more expensive than Qwen3-Next-80B-A3B-Thinking ($0.15/1M tokens).

For output processing, Qwen3.5-397B-A17B ($3.60/1M tokens) is 2.4x more expensive than Qwen3-Next-80B-A3B-Thinking ($1.50/1M tokens).

In conclusion, Qwen3.5-397B-A17B is more expensive than Qwen3-Next-80B-A3B-Thinking.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Mon Jun 22 2026 • llm-stats.com
Alibaba Cloud / Qwen Team
Qwen3.5-397B-A17B
Input tokens$0.60
Output tokens$3.60
Best providerNovita
Alibaba Cloud / Qwen Team
Qwen3-Next-80B-A3B-Thinking
Input tokens$0.15
Output tokens$1.50
Best providerNovita
Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

317.0B diff

Qwen3.5-397B-A17B has 317.0B more parameters than Qwen3-Next-80B-A3B-Thinking, making it 396.3% larger.

Alibaba Cloud / Qwen Team
Qwen3.5-397B-A17B
397.0Bparameters
Alibaba Cloud / Qwen Team
Qwen3-Next-80B-A3B-Thinking
80.0Bparameters
397.0B
Qwen3.5-397B-A17B
80.0B
Qwen3-Next-80B-A3B-Thinking

Context Window

Maximum input and output token capacity

Qwen3.5-397B-A17B accepts 262,144 input tokens compared to Qwen3-Next-80B-A3B-Thinking's 65,536 tokens. Qwen3-Next-80B-A3B-Thinking can generate longer responses up to 65,536 tokens, while Qwen3.5-397B-A17B is limited to 64,000 tokens.

Alibaba Cloud / Qwen Team
Qwen3.5-397B-A17B
Input262,144 tokens
Output64,000 tokens
Alibaba Cloud / Qwen Team
Qwen3-Next-80B-A3B-Thinking
Input65,536 tokens
Output65,536 tokens
Mon Jun 22 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Qwen3.5-397B-A17B supports multimodal inputs, whereas Qwen3-Next-80B-A3B-Thinking does not.

Qwen3.5-397B-A17B can handle both text and other forms of data like images, making it suitable for multimodal applications.

Qwen3.5-397B-A17B

Text
Images
Audio
Video

Qwen3-Next-80B-A3B-Thinking

Text
Images
Audio
Video

License

Usage and distribution terms

Both models are licensed under Apache 2.0.

Both models share the same licensing terms, providing consistent usage rights.

Qwen3.5-397B-A17B

Apache 2.0

Open weights

Qwen3-Next-80B-A3B-Thinking

Apache 2.0

Open weights

Release Timeline

When each model was launched

Qwen3.5-397B-A17B was released on 2026-02-16, while Qwen3-Next-80B-A3B-Thinking was released on 2025-09-10.

Qwen3.5-397B-A17B is 5 months newer than Qwen3-Next-80B-A3B-Thinking.

Qwen3.5-397B-A17B

Feb 16, 2026

4 months ago

5mo newer
Qwen3-Next-80B-A3B-Thinking

Sep 10, 2025

9 months ago

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Provider Availability

Qwen3.5-397B-A17B is available from Novita. Qwen3-Next-80B-A3B-Thinking is available from Novita.

Qwen3.5-397B-A17B

novita logo
Novita
Input Price:Input: $0.60/1MOutput Price:Output: $3.60/1M

Qwen3-Next-80B-A3B-Thinking

novita logo
Novita
Input Price:Input: $0.15/1MOutput Price:Output: $1.50/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

Alibaba Cloud / Qwen Team

Qwen3.5-397B-A17B

View details

Alibaba Cloud / Qwen Team

Larger context window (262,144 tokens)
Supports multimodal inputs
Higher GPQA score (88.4% vs 77.2%)
Higher HMMT25 score (92.7% vs 73.9%)
Higher IFEval score (92.6% vs 88.9%)
Higher Include score (85.6% vs 78.9%)
Higher LiveCodeBench v6 score (83.6% vs 68.7%)
Higher MMLU-Pro score (87.8% vs 82.7%)
Higher MMLU-ProX score (84.7% vs 78.7%)
Higher MMLU-Redux score (94.9% vs 92.5%)
Higher PolyMATH score (73.3% vs 56.3%)
Higher SuperGPQA score (70.4% vs 60.8%)
Alibaba Cloud / Qwen Team
Less expensive input tokens
Less expensive output tokens

Detailed Comparison

AI Model Comparison Table
Feature
Alibaba Cloud / Qwen Team
Qwen3.5-397B-A17B
Alibaba Cloud / Qwen Team
Qwen3-Next-80B-A3B-Thinking

FAQ

Common questions about Qwen3.5-397B-A17B vs Qwen3-Next-80B-A3B-Thinking.

Which is better, Qwen3.5-397B-A17B or Qwen3-Next-80B-A3B-Thinking?

Qwen3.5-397B-A17B significantly outperforms across most benchmarks. Qwen3.5-397B-A17B is made by Alibaba Cloud / Qwen Team and Qwen3-Next-80B-A3B-Thinking is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Qwen3.5-397B-A17B compare to Qwen3-Next-80B-A3B-Thinking in benchmarks?

Qwen3.5-397B-A17B scores MMLU-Redux: 94.9%, HMMT 2025: 94.8%, C-Eval: 93.0%, HMMT25: 92.7%, IFEval: 92.6%. Qwen3-Next-80B-A3B-Thinking scores MMLU-Redux: 92.5%, IFEval: 88.9%, AIME 2025: 87.8%, WritingBench: 84.6%, MMLU-Pro: 82.7%.

Is Qwen3.5-397B-A17B cheaper than Qwen3-Next-80B-A3B-Thinking?

Qwen3-Next-80B-A3B-Thinking is 4.0x cheaper for input tokens. Qwen3.5-397B-A17B costs $0.60/M input and $3.60/M output via novita. Qwen3-Next-80B-A3B-Thinking costs $0.15/M input and $1.50/M output via novita.

What are the context window sizes for Qwen3.5-397B-A17B and Qwen3-Next-80B-A3B-Thinking?

Qwen3.5-397B-A17B supports 262K tokens and Qwen3-Next-80B-A3B-Thinking supports 66K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Qwen3.5-397B-A17B and Qwen3-Next-80B-A3B-Thinking?

Key differences include context window (262K vs 66K), input pricing ($0.60 vs $0.15/M), multimodal support (yes vs no). See the full comparison above for benchmark-by-benchmark results.

Related comparisons