VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Refresh Models —Deals —Regulatory —Sources —

Gemini 1.5 Pro vs Kimi K2 Instruct

20 SHARED BENCHMARKS

Across 20 shared benchmarks, Gemini 1.5 Pro scores higher on 3 and Kimi K2 Instruct on 17. The widest gap is aider_polyglot, where Kimi K2 Instruct scores 60 against 16.9.

GOOGLEVSMOONSHOT20 SHARED3–17 HEAD-TO-HEAD
BenchmarkGemini 1.5 ProMARGINKimi K2 Instruct
AA Intelligence1024
aider_polyglot16.960
AIR-Bench 202482.874.1
anthropic_red_team99.999.3
Artificial Analysis Coding Index23.625.9
bbq94.594.9
GPQA Diamond59.177
GSM8K95.297.3
harmbench79.997.4
HLE4.97.9
humaneval86.694.5
ifeval89.489.8
LiveCodeBench30.561
mmlu86.889.5
MMLU-Pro75.882.5
scicode29.531
simple_safety_tests97.5100
simpleqa23.431
SWE-bench Verified34.269.2
xstest98.898.2

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.

Compare Gemini 1.5 Pro withEVERY TRACKED PAIRING

▶Anthropic12 PAIRINGS
Claude Opus 4.8Claude Opus 4.6Claude Sonnet 4.5Claude 3.5 SonnetClaude Opus 4.5Claude Fable 5Claude Opus 5Claude Opus 4.7Claude Sonnet 4Claude Sonnet 4.6Claude Sonnet 5Claude Opus 4
▶OpenAI11 PAIRINGS
GPT-5.5GPT-4oGPT-5.6 SolGPT-5.2GPT-5.4GPT-5o3GPT-5.1o4-minio1GPT-4o mini
▶Google10 PAIRINGS
Gemini 3.1 ProGemini 3 ProGemma 4 26B A4BGemma 4 31BGemini 2.5 ProGemini 3.5 FlashGemini 3 Flash PreviewGemini 3.6 FlashGemma 4 E2BGemini 2.5 Flash
▶Alibaba11 PAIRINGS
Qwen3.5 35B A3BQwen3.5 397B A17BQwen3 VL 4B (Reasoning)Qwen2.5 Instruct 72BQwen3.5 27BQwen3.6 PlusQwen3.6 27BQwen3.6 35B A3BQwen3.7 MaxQwen3.5 122B A10BQwen3.7 Plus Preview
▶DeepSeek5 PAIRINGS
DeepSeek-V3DeepSeek-V4-ProDeepSeek-V3.2DeepSeek-V4-FlashDeepSeek-R1
▶Moonshot3 PAIRINGS
Kimi K2.5Kimi K2.6Kimi K3
▶Meta2 PAIRINGS
Llama 3.1 Instruct 405BMuse Spark 1.1
▶Z.ai2 PAIRINGS
GLM-5.1GLM-5.2
▶MiniMax1 PAIRING
MiniMax M2.7
▶NVIDIA1 PAIRING
Nemotron 3 Super 120B A12B

Compare Kimi K2 Instruct withEVERY TRACKED PAIRING

▶Anthropic12 PAIRINGS
Claude Opus 4.8Claude Opus 4.6Claude Sonnet 4.5Claude 3.5 SonnetClaude Opus 4.5Claude Fable 5Claude Opus 5Claude Opus 4.7Claude Sonnet 4Claude Sonnet 4.6Claude Sonnet 5Claude Opus 4
▶OpenAI11 PAIRINGS
GPT-5.5GPT-4oGPT-5.6 SolGPT-5.2GPT-5.4GPT-5o3GPT-5.1o4-minio1GPT-4o mini
▶Google10 PAIRINGS
Gemini 3.1 ProGemini 3 ProGemma 4 26B A4BGemma 4 31BGemini 2.5 ProGemini 3.5 FlashGemini 3 Flash PreviewGemini 3.6 FlashGemma 4 E2BGemini 2.5 Flash
▶Alibaba11 PAIRINGS
Qwen3.5 35B A3BQwen3.5 397B A17BQwen3 VL 4B (Reasoning)Qwen2.5 Instruct 72BQwen3.5 27BQwen3.6 PlusQwen3.6 27BQwen3.6 35B A3BQwen3.7 MaxQwen3.5 122B A10BQwen3.7 Plus Preview
▶DeepSeek5 PAIRINGS
DeepSeek-V3DeepSeek-V4-ProDeepSeek-V3.2DeepSeek-V4-FlashDeepSeek-R1
▶Moonshot3 PAIRINGS
Kimi K2.5Kimi K2.6Kimi K3
▶Meta2 PAIRINGS
Llama 3.1 Instruct 405BMuse Spark 1.1
▶Z.ai2 PAIRINGS
GLM-5.1GLM-5.2
▶MiniMax1 PAIRING
MiniMax M2.7
▶NVIDIA1 PAIRING
Nemotron 3 Super 120B A12B