The AI arena is free today

Open Superagent

Hy3 vs Hy4 flash

Hy3 and Hy4 flash are closely matched at 41.5 and 41.0 on the LLM Stats Score. Hy4 flash is 2.0x cheaper per token.

Tencent · Tencent · Updated for 2026

Which is better?

Hy3 and Hy4 flash are closely matched on the overall LLM Stats Score at 41.5 and 41.0.

In the 1 individual benchmarks reported for both models, Hy4 flash wins 1; this is a narrower head-to-head signal than the composite indexes.

On price, Hy4 flash is roughly 2.0x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Hy4 flash also accepts a larger context window (960,000 input tokens), making it the stronger choice for long documents and large codebases.

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

Choose Hy3

  • you need open weights you can self-host or fine-tune

Choose Hy4 flash

  • you value its reported benchmark strengths — it wins 1 of 1 exact shared results
  • cost matters — it's about 2.0x cheaper per token
  • you process long inputs — it offers a 960,000 token context window
  • you want the most recent training data — it shipped Nov 2026

At a glance

The differences that matter most.

Core performance indexes
41.5
#78
41.0
#85
41.8
#71
40.7
#78
30.6
#67
24.0
#106
25.5
#81
27.8
#71
Cost, coverage & limits
Benchmark wins
0 of 1
1 of 1
Input price
$0.14 / M
$0.07 / M
Output price
$0.58 / M
$0.28 / M
Context window
262,144
960,000

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Hy3
Hy4 flash
19.3#92
21.3#78
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

31 reported for Hy3 · 9 for Hy4 flash

1 shared

Hy3 outperforms in 0 benchmarks, while Hy4 flash is better at 1 benchmark (APEX-Agents).

Hy4 flash significantly outperforms across most benchmarks.

Sun Nov 22 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Hy4 flash costs less

For input processing, Hy3 ($0.14/1M tokens) is 2.0x more expensive than Hy4 flash ($0.07/1M tokens).

For output processing, Hy3 ($0.58/1M tokens) is 2.1x more expensive than Hy4 flash ($0.28/1M tokens).

In conclusion, Hy3 is more expensive than Hy4 flash.*

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

Lowest available price from all providers
Sun Nov 22 2026 • llm-stats.com
Tencent
Hy3
Input tokens$0.14
Output tokens$0.58
Best providerDeepinfra
Tencent
Hy4 flash
Input tokens$0.07
Output tokens$0.28
Best providerTencent
Notice missing or incorrect data?

Model Size

Parameter count comparison

71.0B diff

Hy3 has 71.0B more parameters than Hy4 flash, making it 31.7% larger.

Tencent
Hy3
295.0Bparameters
Tencent
Hy4 flash
224.0Bparameters
295.0B
Hy3
224.0B
Hy4 flash

Context Window

Maximum input and output token capacity

Hy4 flash accepts 960,000 input tokens compared to Hy3's 262,144 tokens. Hy3 can generate longer responses up to 262,144 tokens, while Hy4 flash is limited to 64,000 tokens.

Tencent
Hy3
Input262,144 tokens
Output262,144 tokens
Tencent
Hy4 flash
Input960,000 tokens
Output64,000 tokens
Sun Nov 22 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Hy4 flash supports multimodal inputs, whereas Hy3 does not.

Hy4 flash can handle both text and other forms of data like images, making it suitable for multimodal applications.

Hy3

Text
Images
Audio
Video

Hy4 flash

Text
Images
Audio
Video

Release Timeline

When each model was launched

Hy3 was released on 2026-07-06, while Hy4 flash was released on 2026-11-16.

Hy4 flash is 4 months newer than Hy3.

Hy3

Jul 6, 2026

4 months ago

Hy4 flash

Nov 16, 2026

6 days ago

4mo 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

Provider Availability

Hy3 is available from DeepInfra. Hy4 flash is available from Tencent Cloud.

Hy3

deepinfra logo
Deepinfra
Input Price:Input: $0.14/1MOutput Price:Output: $0.58/1M

Hy4 flash

tencent logo
Tencent
Input Price:Input: $0.07/1MOutput Price:Output: $0.28/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?

Judge for yourself.

Run your own prompts against Hy3 and Hy4 flash side-by-side, then vote on the output you prefer.

Hy3
✓ Preferred
Hy4 flash
Open in Playground

FAQ

Common questions about Hy3 vs Hy4 flash.

Which is better, Hy3 or Hy4 flash?

Hy3 and Hy4 flash are closely matched on the LLM Stats Score at 41.5 and 41.0. Hy3 is made by Tencent and Hy4 flash is made by Tencent. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Hy3 compare to Hy4 flash in benchmarks?

Hy3 scores DeepSearchQA: 91.0%, GPQA: 90.4%, IMO-AnswerBench: 90.0%, BrowseComp: 84.2%, MCP Atlas: 79.1%. Hy4 flash scores MMMU-Pro: 75.8%, OSWorld 2.1 (partial): 68.1%, DeepSWE 1.1: 64.6%, APEX-Agents: 63.1%, Humanity's Last Exam (with tools): 55.2%.

Is Hy3 cheaper than Hy4 flash?

Hy4 flash is 2.0x cheaper for input tokens. Hy3 costs $0.14/M input and $0.58/M output via deepinfra. Hy4 flash costs $0.07/M input and $0.28/M output via tencent.

What are the context window sizes for Hy3 and Hy4 flash?

Hy3 supports 262K tokens and Hy4 flash supports 960K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Hy3 and Hy4 flash?

Key differences include LLM Stats Score (41.5 vs 41.0), context window (262K vs 960K), input pricing ($0.14 vs $0.07/M), multimodal support (no vs yes), licensing (Apache 2.0 vs Unknown). See the full comparison above for benchmark-by-benchmark results.