Model comparison

Hy3 vs Mistral Medium 3.5

Hy3 is the stronger model overall, scoring 44.2 to 40.2 on the Noometry Index.

Last verified October 10, 2026. 18 shared benchmarks.

Hy3

Tencent

44.2

Rank #79 Confirmed

Mistral Medium 3.5

Mistral AI

40.2

Rank #152 Confirmed

Summary

They share 18 benchmarks with published results for both. Hy3 scores higher in 8 categories and Mistral Medium 3.5 in 0 categories; 5 gaps are clear of the uncertainty.
The widest gap is in coding, where Hy3 leads 46.8 to 36.0.
The biggest single-benchmark swing is NYT Connections (extended): 41.2% for Hy3 and 12.9% for Mistral Medium 3.5.
Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium 3.5.

Side by side

Hy3 and Mistral Medium 3.5 specifications
	Hy3	Mistral Medium 3.5
Provider	Tencent	Mistral AI
Noometry Index	44.2	40.2
Released	2026-07-06	—
Weights	Open	Open
Context window	262K	262K
Max output	128K	210K
Input $ / M tokens	$0.0825	$1.50
Output $ / M tokens	$0.33	$7.50
Results tracked	19	22

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Hy3 leads

Hy3: 46.8 (#63), Mistral Medium 3.5: 36.0 (#213)

Coding benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena WebDev	1508	1264
LMArena Coding	1464	1461

Reasoning Hy3 leads

Hy3: 26.1 (#136), Mistral Medium 3.5: 17.3 (#295)

Reasoning benchmarks
Benchmark	Hy3	Mistral Medium 3.5
NYT Connections (extended)	41.2%	12.9%
LMArena Hard Prompts	1447	1436
Kagi LLM Benchmark	—	41.4%
Epoch Capabilities Index	—	141.35

Math Hy3 leads

Hy3: 40.1 (#93), Mistral Medium 3.5: 39.1 (#113)

Math benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Math	1475	1431

Knowledge Too close to call

Hy3: 40.8 (#114), Mistral Medium 3.5: 40.0 (#126)

Knowledge benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Expert	1460	1432

Multimodal Not comparable

Hy3: —, Mistral Medium 3.5: 38.3 (#65)

Multimodal benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Vision	—	1223

Multilingual Hy3 leads

Hy3: 53.5 (#65), Mistral Medium 3.5: 51.9 (#100)

Multilingual benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Non-English	1426	1404
LMArena Chinese	1493	1442
LMArena French	1461	1448
LMArena German	1439	1451
LMArena Korean	1395	1385
LMArena Russian	1432	1395
LMArena Spanish	1456	1409
LMArena Japanese	1392	—

Instruction Following Too close to call

Hy3: 75.1 (#70), Mistral Medium 3.5: 74.6 (#90)

Instruction Following benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Instruction Following	1426	1415

Long Context Too close to call

Hy3: 44.1 (#75), Mistral Medium 3.5: 43.2 (#103)

Long Context benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Longer Query	1442	1415

Writing & Preference Hy3 leads

Hy3: 62.2 (#81), Mistral Medium 3.5: 58.5 (#117)

Writing & Preference benchmarks
Benchmark	Hy3	Mistral Medium 3.5
LMArena Text	1439	1421
LMArena Creative Writing	1402	1374
LMArena Multi-Turn	1436	1423
EQ-Bench 4	—	993

Frequently asked questions

Is Hy3 better than Mistral Medium 3.5?

Hy3 is the stronger model overall, scoring 44.2 to 40.2 on the Noometry Index.

Which is cheaper, Hy3 or Mistral Medium 3.5?

Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Mistral Medium 3.5 lists at $1.50 and $7.50.

Is Hy3 or Mistral Medium 3.5 better for coding?

Hy3 scores higher on coding benchmarks: 46.8 versus 36.0 in the Noometry coding category.

Which has the bigger context window?

Both accept 262K tokens.

How many benchmarks do Hy3 and Mistral Medium 3.5 share?

18 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Mistral Medium 3.5 has 22.