Model comparison

Hunyuan Turbos 20250226 vs Phi 3 Mini 4k Instruct

Hunyuan Turbos 20250226 is the stronger model overall, scoring 40.6 to 27.4 on the Noometry Index.

Last verified . 16 shared benchmarks.

Hunyuan Turbos 20250226 Tencent

40.6

Rank #138 Confirmed

Phi 3 Mini 4k Instruct Microsoft

27.4

Rank #324 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Hunyuan Turbos 20250226 scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Hunyuan Turbos 20250226 leads 57.5 to 27.7.
  • Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.

Side by side

Hunyuan Turbos 20250226 and Phi 3 Mini 4k Instruct specifications
Hunyuan Turbos 20250226Phi 3 Mini 4k Instruct
ProviderTencentMicrosoft
Noometry Index40.627.4
Released—2024-04-23
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked1635

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

Category by category

Coding Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 39.3 (#148), Phi 3 Mini 4k Instruct: 26.1 (#319)

Coding benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Coding13611093
LiveBench Coding—15.5%
HumanEval+—59.1%
MBPP+—54.2%

Reasoning Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 26.7 (#115), Phi 3 Mini 4k Instruct: 13.6 (#326)

Reasoning benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Hard Prompts13741072
Chess Puzzles—0%
LiveBench Reasoning—26.8%
LiveBench Data Analysis—34.7%
Adversarial NLI—52.8%
BIG-Bench Hard—71.7%
HellaSwag—76.7%
LiveBench—22.4%
WinoGrande—70.8%

Math Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 34.8 (#155), Phi 3 Mini 4k Instruct: 24.5 (#258)

Math benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Math13591111
LiveBench Math—15.7%

Knowledge Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 36.4 (#158), Phi 3 Mini 4k Instruct: 27.9 (#241)

Knowledge benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Expert13381045
ARC (AI2) Challenge—84.9%
MMLU—68.8%
OpenBookQA—88%
TriviaQA—64%

Multilingual Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 49.1 (#132), Phi 3 Mini 4k Instruct: 26.5 (#275)

Multilingual benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Non-English13621021
LMArena Chinese14161021
LMArena French13901075
LMArena German13561044
LMArena Japanese1338935
LMArena Korean1351905
LMArena Russian13681022
LMArena Spanish—1084

Instruction Following Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 71.2 (#153), Phi 3 Mini 4k Instruct: 48.0 (#298)

Instruction Following benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Instruction Following13441053
LiveBench Instruction Following—39.1%

Long Context Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 41.6 (#132), Phi 3 Mini 4k Instruct: 31.7 (#271)

Long Context benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Longer Query13651044

Writing & Preference Hunyuan Turbos 20250226 leads

Hunyuan Turbos 20250226: 57.5 (#125), Phi 3 Mini 4k Instruct: 27.7 (#295)

Writing & Preference benchmarks
BenchmarkHunyuan Turbos 20250226Phi 3 Mini 4k Instruct
LMArena Text13771073
LMArena Creative Writing13581037
LMArena Multi-Turn13871018
LiveBench Language—9.2%

Frequently asked questions

Is Hunyuan Turbos 20250226 better than Phi 3 Mini 4k Instruct?

Hunyuan Turbos 20250226 is the stronger model overall, scoring 40.6 to 27.4 on the Noometry Index.

Is Hunyuan Turbos 20250226 or Phi 3 Mini 4k Instruct better for coding?

Hunyuan Turbos 20250226 scores higher on coding benchmarks: 39.3 versus 26.1 in the Noometry coding category.

How many benchmarks do Hunyuan Turbos 20250226 and Phi 3 Mini 4k Instruct share?

16 benchmarks have published results for both models. Hunyuan Turbos 20250226 has 16 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.

Related comparisons

Go deeper