Noometry

Model comparison

Hy3 vs Llama 3.1-8B

Hy3 is the stronger model overall, scoring 43.6 to 22.8 on the Noometry Index. Llama 3.1-8B costs 4.0× less per token, which makes it the better buy when Hy3's lead doesn't matter for your workload.

Last verified . 15 shared benchmarks.

Hy3 Tencent

43.6

Rank #79 Confirmed

Llama 3.1-8B Meta

22.8

Rank #347 Confirmed

Summary

  • They share 15 benchmarks with published results for both. Hy3 scores higher in 8 categories and Llama 3.1-8B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Hy3 leads 62.5 to 29.9.
  • Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.13 / $0.53 for Hy3.
  • Hy3 accepts more context: 262K tokens versus 128K.

Side by side

Hy3Llama 3.1-8B
ProviderTencentMeta
Noometry Index43.622.8
Released2026-07-062024-07-23
WeightsOpenOpen
Context window262K128K
Max output128K4K
Input $ / M tokens$0.13$0.05
Output $ / M tokens$0.53$0.08
Results tracked1743

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

Category by category

Coding Hy3 leads

Hy3: 46.6 (#60), Llama 3.1-8B: 19.9 (#335)

BenchmarkHy3Llama 3.1-8B
LMArena Coding14651195
LMArena WebDev1509—
SciCode—13.2%
WeirdML—1.7%
BigCodeBench Instruct—32.8%
BigCodeBench Complete—40.5%
HumanEval+—62.8%
MBPP+—55.6%

Agentic & Tool Use Not comparable

Hy3: —, Llama 3.1-8B: 22.3 (#123)

BenchmarkHy3Llama 3.1-8B
Berkeley Function Calling Leaderboard—25.8%
BALROG—15.1%

Reasoning Hy3 leads

Hy3: 25.0 (#134), Llama 3.1-8B: 14.5 (#316)

BenchmarkHy3Llama 3.1-8B
LMArena Hard Prompts14491175
NYT Connections (extended)41.2%—
CritPt—0%
Chess Puzzles—0%
DTBench—50.9%
LMCA—5.4%
Epoch Capabilities Index—116.48
PIQA—81.2%

Math Hy3 leads

Hy3: 37.4 (#98), Llama 3.1-8B: 9.5 (#311)

BenchmarkHy3Llama 3.1-8B
LMArena Math14751179
OTIS Mock AIME 2024-2025—1.7%
Omni-MATH—13.7%
MATH Level 5—22.9%
GSM8K—82.4%

Knowledge Hy3 leads

Hy3: 40.2 (#108), Llama 3.1-8B: 8.0 (#302)

BenchmarkHy3Llama 3.1-8B
LMArena Expert14611144
GPQA Diamond—27%
MMLU-Pro—40.6%
GPQA (HELM)—24.7%
BoolQ—82.8%
MMLU—56.1%

Multilingual Hy3 leads

Hy3: 53.7 (#60), Llama 3.1-8B: 34.2 (#244)

BenchmarkHy3Llama 3.1-8B
LMArena Non-English14271148
LMArena Chinese14961151
LMArena French14611176
LMArena Korean13901053
LMArena Russian14351158
LMArena Spanish14481169
LMArena German—1144
LMArena Japanese—1061

Instruction Following Hy3 leads

Hy3: 75.3 (#68), Llama 3.1-8B: 59.1 (#253)

BenchmarkHy3Llama 3.1-8B
LMArena Instruction Following14251159
IFEval—74.3%

Long Context Hy3 leads

Hy3: 44.1 (#70), Llama 3.1-8B: 35.8 (#233)

BenchmarkHy3Llama 3.1-8B
LMArena Longer Query14431182

Writing & Preference Hy3 leads

Hy3: 62.5 (#75), Llama 3.1-8B: 29.9 (#285)

BenchmarkHy3Llama 3.1-8B
LMArena Text14421187
LMArena Creative Writing13991154
LMArena Multi-Turn14361172
EQ-Bench Creative Writing—713
WildBench—68.7%

Frequently asked questions

Is Hy3 better than Llama 3.1-8B?

Hy3 is the stronger model overall, scoring 43.6 to 22.8 on the Noometry Index. Llama 3.1-8B costs 4.0× less per token, which makes it the better buy when Hy3's lead doesn't matter for your workload.

Which is cheaper, Hy3 or Llama 3.1-8B?

Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Hy3 lists at $0.13 and $0.53.

Is Hy3 or Llama 3.1-8B better for coding?

Hy3 scores higher on coding benchmarks: 46.6 versus 19.9 in the Noometry coding category.

Which has the bigger context window?

Hy3 does, with 262K tokens against 128K.

How many benchmarks do Hy3 and Llama 3.1-8B share?

15 benchmarks have published results for both models. Hy3 has 17 scored results on Noometry and Llama 3.1-8B has 43.

Related comparisons

Go deeper