GPTCrunch

Compare Models

Select up to 4 models to compare benchmarks, pricing, and capabilities side by side.

OpenAI logoGPT-4o

OpenAI

Moonshot AI logoKimi K2.5

Moonshot AI

NVIDIA logoLlama 3.1 Nemotron 70B

NVIDIA

Add Model
MMLU
GPT-4o
88.7
Kimi K2.5
92.0
Llama 3.1 Nemotron 70B
86.0
HumanEval
GPT-4o
90.2
Kimi K2.5
99.0
Llama 3.1 Nemotron 70B
85.0
GSM8K
GPT-4o
95.8
Kimi K2.5
99.0
Llama 3.1 Nemotron 70B
94.0
GPQA
GPT-4o
53.6
Kimi K2.5
87.6
Llama 3.1 Nemotron 70B
50.0
MGSM
GPT-4o
90.5
Kimi K2.5
96.0
Llama 3.1 Nemotron 70B
0.0
ARC-Challenge
GPT-4o
96.7
Kimi K2.5
0.0
Llama 3.1 Nemotron 70B
95.0
HellaSwag
GPT-4o
95.3
Kimi K2.5
0.0
Llama 3.1 Nemotron 70B
89.0
MATH
GPT-4o
76.6
Kimi K2.5
98.0
Llama 3.1 Nemotron 70B
0.0
SWE-bench
GPT-4o
38.4
Kimi K2.5
76.8
Llama 3.1 Nemotron 70B
0.0
MMMLU
GPT-4o
85.1
Kimi K2.5
0.0
Llama 3.1 Nemotron 70B
0.0
LiveCodeBench
GPT-4o
0.0
Kimi K2.5
85.0
Llama 3.1 Nemotron 70B
0.0
IFEval
GPT-4o
0.0
Kimi K2.5
94.0
Llama 3.1 Nemotron 70B
92.0
AIME 2025
GPT-4o
0.0
Kimi K2.5
96.1
Llama 3.1 Nemotron 70B
0.0
ModelInputOutputBlended*
GPT-4o
$2.50$10.00$6.25
Kimi K2.5
$0.45$2.20$1.33
Llama 3.1 Nemotron 70B
$0.35$1.05$0.70

*Blended = average of input and output price

Spec
GPT-4o
Kimi K2.5
Llama 3.1 Nemotron 70B
Context Window128K256K128K
Max Output16K16KN/A
TTFT320ms500msN/A
Speed95 tok/s70 tok/sN/A
Parameters~1.8T (estimated)1T (32B active)70B
ArchitectureTransformer (MoE)MoE + MultimodalDense Transformer
Open SourceNoNoYes
Tierfrontierfrontierfrontier

Quick Verdict

Best Performance

Kimi K2.5

Best Value

Llama 3.1 Nemotron 70B

Fastest

GPT-4o