Model Comparison

DeepSeek-V4-Pro-Max vs Hy3Which is better in 2026?

Hy3 has a slight edge in benchmark performance.

Verdict: DeepSeek-V4-Pro-Max vs Hy3 — which is better?

DeepSeek-V4-Pro-Max (by DeepSeek) and Hy3 (by Tencent) 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.

DeepSeek-V4-Pro-Max outperforms in 4 benchmarks (MathArena Apex, SWE-bench Multilingual, SWE-Bench Verified, Toolathlon), while Hy3 is better at 5 benchmarks (BrowseComp, GPQA, IMO-AnswerBench, MCP Atlas, SWE-Bench Pro). Hy3 has a slight edge in benchmark performance.

Choose DeepSeek-V4-Pro-Max if…

  • you want predictable pricing at $1.60/M input and $3.20/M output

Choose Hy3 if…

  • you want the strongest raw capability — it leads on 5 of 9 shared benchmarks
  • you want the most recent training data — it shipped Jul 2026

Performance Benchmarks

Comparative analysis across standard metrics

9 benchmarks

DeepSeek-V4-Pro-Max outperforms in 4 benchmarks (MathArena Apex, SWE-bench Multilingual, SWE-Bench Verified, Toolathlon), while Hy3 is better at 5 benchmarks (BrowseComp, GPQA, IMO-AnswerBench, MCP Atlas, SWE-Bench Pro).

Hy3 has a slight edge in benchmark performance.

Fri Jul 17 2026 • llm-stats.com

Arena Performance

Human preference votes

Model Size

Parameter count comparison

1305.0B diff

DeepSeek-V4-Pro-Max has 1305.0B more parameters than Hy3, making it 442.4% larger.

DeepSeek
DeepSeek-V4-Pro-Max
1.6Tparameters
Tencent
Hy3
295.0Bparameters
1600.0B
DeepSeek-V4-Pro-Max
295.0B
Hy3

Context Window

Maximum input and output token capacity

Only DeepSeek-V4-Pro-Max specifies input context (1,048,576 tokens). Only DeepSeek-V4-Pro-Max specifies output context (131,072 tokens).

DeepSeek
DeepSeek-V4-Pro-Max
Input1,048,576 tokens
Output131,072 tokens
Tencent
Hy3
Input- tokens
Output- tokens
Fri Jul 17 2026 • llm-stats.com

License

Usage and distribution terms

DeepSeek-V4-Pro-Max is licensed under MIT, while Hy3 uses Apache 2.0.

License differences may affect how you can use these models in commercial or open-source projects.

DeepSeek-V4-Pro-Max

MIT

Open weights

Hy3

Apache 2.0

Open weights

Release Timeline

When each model was launched

DeepSeek-V4-Pro-Max was released on 2026-04-23, while Hy3 was released on 2026-07-06.

Hy3 is 2 months newer than DeepSeek-V4-Pro-Max.

DeepSeek-V4-Pro-Max

Apr 23, 2026

2 months ago

Hy3

Jul 6, 2026

1 weeks ago

2mo newer

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

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

DeepSeek
Larger context window (1,048,576 tokens)
Higher MathArena Apex score (90.2% vs 38.7%)
Higher SWE-bench Multilingual score (76.2% vs 75.8%)
Higher SWE-Bench Verified score (80.6% vs 78.0%)
Higher Toolathlon score (51.8% vs 48.5%)
Tencent
Higher BrowseComp score (84.2% vs 83.4%)
Higher GPQA score (90.4% vs 90.1%)
Higher IMO-AnswerBench score (90.0% vs 89.8%)
Higher MCP Atlas score (79.1% vs 73.6%)
Higher SWE-Bench Pro score (57.9% vs 55.4%)

Detailed Comparison

Interactive Arena

Judge for yourself.

Run your own prompts against DeepSeek-V4-Pro-Max and Hy3 side-by-side, then vote on the output you prefer.

DeepSeek-V4-Pro-Max
✓ Preferred
Hy3
Open in Playground
AI Model Comparison Table
Feature
DeepSeek
DeepSeek-V4-Pro-Max
Tencent
Hy3

FAQ

Common questions about DeepSeek-V4-Pro-Max vs Hy3.

Which is better, DeepSeek-V4-Pro-Max or Hy3?

Hy3 has a slight edge in benchmark performance. DeepSeek-V4-Pro-Max is made by DeepSeek and Hy3 is made by Tencent. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does DeepSeek-V4-Pro-Max compare to Hy3 in benchmarks?

DeepSeek-V4-Pro-Max scores CodeForces: 100.0%, HMMT Feb 26: 95.2%, LiveCodeBench: 93.5%, MathArena Apex: 90.2%, GPQA: 90.1%. Hy3 scores DeepSearchQA: 91.0%, GPQA: 90.4%, IMO-AnswerBench: 90.0%, BrowseComp: 84.2%, MCP Atlas: 79.1%.

What are the context window sizes for DeepSeek-V4-Pro-Max and Hy3?

DeepSeek-V4-Pro-Max supports 1.0M tokens and Hy3 supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between DeepSeek-V4-Pro-Max and Hy3?

Key differences include licensing (MIT vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes DeepSeek-V4-Pro-Max and Hy3?

DeepSeek-V4-Pro-Max is developed by DeepSeek and Hy3 is developed by Tencent.

Related comparisons