Noometry

Model comparison

GPT-3.5-turbo vs Hy3

Hy3 is the stronger model overall, scoring 43.6 to 23.1 on the Noometry Index.

Last verified . 15 shared benchmarks.

GPT-3.5-turbo OpenAI

23.1

Rank #345 Confirmed

Hy3 Tencent

43.6

Rank #79 Confirmed

Summary

  • They share 15 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Hy3 leads 62.5 to 25.4.
  • Hy3 is cheaper at $0.13 / $0.53 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
  • Hy3 accepts more context: 262K tokens versus 16K.
  • Hy3 has downloadable open weights; the other is API-only.

Side by side

GPT-3.5-turboHy3
ProviderOpenAITencent
Noometry Index23.143.6
Released2023-03-012026-07-06
WeightsProprietaryOpen
Context window16K262K
Max output4K128K
Input $ / M tokens$0.50$0.13
Output $ / M tokens$1.50$0.53
Results tracked4417

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

Category by category

Coding Hy3 leads

GPT-3.5-turbo: 23.5 (#326), Hy3: 46.6 (#60)

BenchmarkGPT-3.5-turboHy3
LMArena Coding11371465
LMArena WebDev—1509
WeirdML3.5%—
BigCodeBench Instruct39.1%—
BigCodeBench Complete50.6%—
HumanEval+70.7%—
MBPP+69.7%—

Agentic & Tool Use Not comparable

GPT-3.5-turbo: —, Hy3: —

BenchmarkGPT-3.5-turboHy3
METR Time Horizons21.5%—

Reasoning Hy3 leads

GPT-3.5-turbo: 13.3 (#328), Hy3: 25.0 (#134)

BenchmarkGPT-3.5-turboHy3
LMArena Hard Prompts11081449
NYT Connections (extended)—41.2%
Chess Puzzles0%—
Mystery Game Puzzles3%—
DTBench48.5%—
LMCA9.7%—
Adversarial NLI58.1%—
BIG-Bench Hard61.6%—
CommonsenseQA 2.057%—
Epoch Capabilities Index118.48—
ForecastBench50.4—
WinoGrande81.6%—

Math Hy3 leads

GPT-3.5-turbo: 6.0 (#321), Hy3: 37.4 (#98)

BenchmarkGPT-3.5-turboHy3
LMArena Math11421475
FrontierMath (Tiers 1-3)0%—
OTIS Mock AIME 2024-20252.2%—
MATH Level 515.9%—
GSM8K57.8%—

Knowledge Hy3 leads

GPT-3.5-turbo: 10.0 (#298), Hy3: 40.2 (#108)

BenchmarkGPT-3.5-turboHy3
LMArena Expert10701461
GPQA Diamond28%—
ARC (AI2) Challenge87.4%—
BoolQ87%—
MMLU71.4%—
OpenBookQA86%—
TriviaQA85.8%—

Multilingual Hy3 leads

GPT-3.5-turbo: 31.6 (#253), Hy3: 53.7 (#60)

BenchmarkGPT-3.5-turboHy3
LMArena Non-English11081427
LMArena Chinese10751496
LMArena French11171461
LMArena Korean10191390
LMArena Russian11231435
LMArena Spanish11201448
LMArena German1090—
LMArena Japanese1043—

Instruction Following Hy3 leads

GPT-3.5-turbo: 58.1 (#257), Hy3: 75.3 (#68)

BenchmarkGPT-3.5-turboHy3
LMArena Instruction Following11191425

Long Context Hy3 leads

GPT-3.5-turbo: 34.0 (#249), Hy3: 44.1 (#70)

BenchmarkGPT-3.5-turboHy3
LMArena Longer Query11211443

Writing & Preference Hy3 leads

GPT-3.5-turbo: 25.4 (#300), Hy3: 62.5 (#75)

BenchmarkGPT-3.5-turboHy3
LMArena Text11251442
LMArena Creative Writing10921399
LMArena Multi-Turn11171436
EQ-Bench Creative Writing451—

Frequently asked questions

Is GPT-3.5-turbo better than Hy3?

Hy3 is the stronger model overall, scoring 43.6 to 23.1 on the Noometry Index.

Which is cheaper, GPT-3.5-turbo or Hy3?

Hy3 is cheaper. It lists at $0.13 per million input tokens and $0.53 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.

Is GPT-3.5-turbo or Hy3 better for coding?

Hy3 scores higher on coding benchmarks: 46.6 versus 23.5 in the Noometry coding category.

Which has the bigger context window?

Hy3 does, with 262K tokens against 16K.

How many benchmarks do GPT-3.5-turbo and Hy3 share?

15 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Hy3 has 17.

Related comparisons

Go deeper