Model Comparison
Qwen3.5-397B-A17B vs Mercury 2Which is better in 2026?
Qwen3.5-397B-A17B significantly outperforms across most benchmarks. Mercury 2 is 3.6x cheaper per token.
Verdict: Qwen3.5-397B-A17B vs Mercury 2 — which is better?
Qwen3.5-397B-A17B (by Alibaba Cloud / Qwen Team) and Mercury 2 (by Inception) 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 2 benchmarks (GPQA, IFBench), while Mercury 2 is better at 0 benchmarks. Qwen3.5-397B-A17B significantly outperforms across most benchmarks.
On price, Mercury 2 is roughly 3.6x 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 2 of 2 shared benchmarks
- you process long inputs — it offers a 262,144 token context window
- you need open weights you can self-host or fine-tune
Choose Mercury 2 if…
- cost matters — it's about 3.6x cheaper per token
- you want the most recent training data — it shipped Feb 2026
Performance Benchmarks
Comparative analysis across standard metrics
Qwen3.5-397B-A17B outperforms in 2 benchmarks (GPQA, IFBench), while Mercury 2 is better at 0 benchmarks.
Qwen3.5-397B-A17B significantly outperforms across most benchmarks.
Arena Performance
Human preference votes
Pricing Analysis
Price comparison per million tokens
For input processing, Qwen3.5-397B-A17B ($0.60/1M tokens) is 2.4x more expensive than Mercury 2 ($0.25/1M tokens).
For output processing, Qwen3.5-397B-A17B ($3.60/1M tokens) is 4.8x more expensive than Mercury 2 ($0.75/1M tokens).
In conclusion, Qwen3.5-397B-A17B is more expensive than Mercury 2.*
* Using a 3:1 ratio of input to output tokens
Context Window
Maximum input and output token capacity
Qwen3.5-397B-A17B accepts 262,144 input tokens compared to Mercury 2's 128,000 tokens. Qwen3.5-397B-A17B can generate longer responses up to 64,000 tokens, while Mercury 2 is limited to 8,192 tokens.
Input Capabilities
Supported data types and modalities
Qwen3.5-397B-A17B supports multimodal inputs, whereas Mercury 2 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
Mercury 2
License
Usage and distribution terms
Qwen3.5-397B-A17B is licensed under Apache 2.0, while Mercury 2 uses a proprietary license.
License differences may affect how you can use these models in commercial or open-source projects.
Apache 2.0
Open weights
Proprietary
Closed source
Release Timeline
When each model was launched
Qwen3.5-397B-A17B was released on 2026-02-16, while Mercury 2 was released on 2026-02-24.
Mercury 2 is 0 month newer than Qwen3.5-397B-A17B.
Feb 16, 2026
3 months ago
Feb 24, 2026
3 months ago
1w newerKnowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
Qwen3.5-397B-A17B is available from Novita. Mercury 2 is available from Inception.
Qwen3.5-397B-A17B
Mercury 2
Outputs Comparison
Key Takeaways
Qwen3.5-397B-A17B
View detailsAlibaba Cloud / Qwen Team
Mercury 2
View detailsInception
Detailed Comparison
| Feature | | |
|---|
FAQ
Common questions about Qwen3.5-397B-A17B vs Mercury 2.