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

Kimi K2 Instruct vs Qwen3 VL 4B (Reasoning)

24 SHARED BENCHMARKS

Across 24 shared benchmarks, Kimi K2 Instruct scores higher on 20 and Qwen3 VL 4B (Reasoning) on 3, with 1 level. The widest gap is τ²-Bench Telecom (AA run), where Kimi K2 Instruct scores 73.4 against 15.5.

MOONSHOTVSALIBABA24 SHARED20–3 HEAD-TO-HEAD
BenchmarkKimi K2 InstructMARGINQwen3 VL 4B (Reasoning)
AA Agentic Index37.714.4
AA Intelligence247.7
AA-LCR53.723
AA-Omniscience-26.6-68.4
AIME 202469.672.9
AIME 20255774.5
Artificial Analysis Coding Index25.96.7
critpt00
gdpval18.213.8
GPQA Diamond7764.1
HLE7.94.6
HMMT 202538.853.1
IFBench4236.6
ifeval89.882.6
LiveCodeBench6151.3
LiveCodeBench v656.151.3
mmlu_redux92.786
MMLU-Pro82.573.6
OmniScience Accuracy25.412.2
OmniScience Non-Hallucination30.38.2
scicode3117.1
supergpqa57.246.8
Terminal-Bench Hard23.51.5
τ²-Bench Telecom (AA run)73.415.5

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 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
▶Google11 PAIRINGS
Gemini 3.1 ProGemini 3 ProGemma 4 26B A4BGemini 1.5 ProGemma 4 31BGemini 2.5 ProGemini 3.5 FlashGemini 3 Flash PreviewGemini 3.6 FlashGemma 4 E2BGemini 2.5 Flash
▶Alibaba10 PAIRINGS
Qwen3.5 35B A3BQwen3.5 397B A17BQwen2.5 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 Qwen3 VL 4B (Reasoning) 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
▶Google11 PAIRINGS
Gemini 3.1 ProGemini 3 ProGemma 4 26B A4BGemini 1.5 ProGemma 4 31BGemini 2.5 ProGemini 3.5 FlashGemini 3 Flash PreviewGemini 3.6 FlashGemma 4 E2BGemini 2.5 Flash
▶Alibaba10 PAIRINGS
Qwen3.5 35B A3BQwen3.5 397B A17BQwen2.5 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