The AI arena is free today

Open Superagent

Hy4 preview vs Kimi K3

Hy4 preview and Kimi K3 are closely matched at 51.5 and 54.6 on the LLM Stats Score.

Tencent · Moonshot AI · Updated for 2026

Which is better?

Hy4 preview and Kimi K3 are closely matched on the overall LLM Stats Score at 51.5 and 54.6.

In the 12 individual benchmarks reported for both models, Kimi K3 wins 8; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose Hy4 preview

  • you want the most recent training data — it shipped Aug 2026

Choose Kimi K3

  • you value its reported benchmark strengths — it wins 8 of 12 exact shared results

At a glance

The differences that matter most.

Hy4 preview
Tencent
Kimi K3
Moonshot AI
Core performance indexes
LLM Stats Score
51.5
#10
54.6
#5
Reasoning
51.2
#8
53.3
#4
Coding
40.9
#10
45.0
#4
Agents
37.1
#11
40.3
#3
Cost, coverage & limits
Benchmark wins
4 of 12
8 of 12
Input price
— / M
$3.00 / M
Output price
— / M
$15.00 / M
Context window
—
1,048,576

Capability indexes

Additional strengths measured across groups of related public benchmarks

2 shared
Index
Hy4 preview
Kimi K3
Math
40.3#12
40.9#10
Tool use
32.6#7
36.1#1
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

32 reported for Hy4 preview · 31 for Kimi K3

12 shared

Hy4 preview outperforms in 4 benchmarks (AutomationBench, Job Bench, OfficeQA Pro, Toolathlon), while Kimi K3 is better at 8 benchmarks (APEX-Agents, DeepSWE, GPQA, MCP Atlas, PostTrainBench, Program Bench, SWE-Marathon, Terminal-Bench 2.1).

Kimi K3 shows notably better performance in the majority of benchmarks.

Mon Aug 31 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

2030.0B diff

Kimi K3 has 2030.0B more parameters than Hy4 preview, making it 263.6% larger.

Tencent
Hy4 preview
770.0Bparameters
Moonshot AI
Kimi K3
2.8Tparameters
770.0B
Hy4 preview
2800.0B
Kimi K3

Context Window

Maximum input and output token capacity

Only Kimi K3 specifies input context (1,048,576 tokens). Only Kimi K3 specifies output context (1,048,576 tokens).

Tencent
Hy4 preview
Input- tokens
Output- tokens
Moonshot AI
Kimi K3
Input1,048,576 tokens
Output1,048,576 tokens
Mon Aug 31 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Kimi K3 supports multimodal inputs, whereas Hy4 preview does not.

Kimi K3 can handle both text and other forms of data like images, making it suitable for multimodal applications.

Hy4 preview

Text
Images
Audio
Video

Kimi K3

Text
Images
Audio
Video

License

Usage and distribution terms

Hy4 preview is licensed under Apache 2.0, while Kimi K3 uses Kimi K3 License.

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

Hy4 preview

Apache 2.0

Open weights

Kimi K3

Kimi K3 License

Open weights

Release Timeline

When each model was launched

Hy4 preview was released on 2026-08-28, while Kimi K3 was released on 2026-07-16.

Hy4 preview is 1 month newer than Kimi K3.

Hy4 preview

Aug 28, 2026

2 days ago

1mo newer
Kimi K3

Jul 16, 2026

1 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

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Hy4 preview and Kimi K3 side-by-side, then vote on the output you prefer.

Hy4 preview
✓ Preferred
Kimi K3
Open in Playground

FAQ

Common questions about Hy4 preview vs Kimi K3.

Which is better, Hy4 preview or Kimi K3?

Hy4 preview and Kimi K3 are closely matched on the LLM Stats Score at 51.5 and 54.6. Hy4 preview is made by Tencent and Kimi K3 is made by Moonshot AI. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Hy4 preview compare to Kimi K3 in benchmarks?

Hy4 preview scores GPQA: 92.3%, Terminal-Bench 2.1: 85.4%, WideSearch: 83.9%, MCP Atlas: 83.7%, SWE-bench Multilingual: 82.9%. Kimi K3 scores MathVision: 97.8%, DeepSearchQA: 95.0%, GPQA: 93.5%, CharXiv-R: 91.3%, BrowseComp: 91.2%.

What are the context window sizes for Hy4 preview and Kimi K3?

Hy4 preview supports an unknown number of tokens and Kimi K3 supports 1.0M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Hy4 preview and Kimi K3?

Key differences include LLM Stats Score (51.5 vs 54.6), multimodal support (no vs yes), licensing (Apache 2.0 vs Kimi K3 License). See the full comparison above for benchmark-by-benchmark results.

Who makes Hy4 preview and Kimi K3?

Hy4 preview is developed by Tencent and Kimi K3 is developed by Moonshot AI.

Related comparisons