// COMPARISON

Kimi K2 Thinking vs Qwen3-30B-A3B

A head-to-head benchmark comparison of Kimi K2 Thinking vs Qwen3-30B-A3B across 48 evaluations. Kimi K2 Thinking leads on the most benchmarks (27).

The models

Leads 27
MO

Kimi K2 Thinking

Moonshot
Language Open (Restricted)
Params
1T
Context
β€”
Released
Nov 2025

Today, we are introducing Kimi K2 Thinking, our best open-source thinking model. Built as a thinking agent, it reasons step by step while using tools, achieving.

Full Kimi K2 Thinking specs β†’
Qwen

Qwen3-30B-A3B

Qwen
Language Open (Restricted)
Params
30B
Context
β€”
Released
Apr 2025

Today, we are excited to announce the release of Qwen3, the latest addition to the Qwen family of large language models. Our flagship model, Qwen3-235B-A22B, ac.

Full Qwen3-30B-A3B specs β†’

Benchmark comparison

Benchmark Kimi K2 Thinking Qwen3-30B-A3B
agentic
BrowseComp 60.2% β€”
BrowseComp-Zh 62.3 β€”
FinSearchComp-T3 47.4 β€”
MCP-Mark 20.4 β€”
MCP-Universe 35.6 β€”
Seal-0 56.3 β€”
Terminal-Bench 2.0 47.1% β€”
Tool-Decathlon 17.6 β€”
τ²-Bench (TAU2) 74.3 β€”
code
Codeforces Elo β€” 1,974
LiveCodeBench 82.6% β€”
LiveCodeBench v5 β€” 62.6%
LiveCodeBench v6 83.1% β€”
SWE-bench Verified 71.3% β€”
SciCode 44.8% 28.5%
coding
Multi-SWE-bench 41.9 β€”
OJ-Bench (cpp) 48.7 β€”
SWE-Bench Multilingual 61.1 β€”
general
AlignBench v1.1 β€” 8.7
ArenaHard β€” 91%
C-Eval β€” 86.6%
Creative Writing v3 β€” 79.1
IFEval β€” 86.5%
LiveBench β€” 74.3
Longform Writing 73.8 β€”
MMLU-Pro 84.6% β€”
MMLU-Redux 94.4% 89.5%
MMMLU β€” 78.4%
Multi-IF β€” 72.2
WritingBench β€” 7.7
knowledge
INCLUDE β€” 71.9
long-context
Frames 87 β€”
math
AIME 2024 β€” 80.4%
AIME 2025 100% 70.9%
HMMT February 2025 97.5% 23.54%
HMMT November 2025 89.2 β€”
IMOAnswerBench 78.6 β€”
MATH-500 β€” 98%
MT-AIME 2024 β€” 73.9
Omni-MATH-HARD β€” 23.87%
PolyMath-en β€” 46.1
reasoning
AutoLogi β€” 88.7
GPQA Diamond 84.5% 65.8%
Humanity's Last Exam 51% 6.14%
MLogiQA β€” 70.1
ZebraLogic β€” 89.5%
safety
HealthBench 58% β€”
tool-use
BFCL v3 β€” 69.1%

Best result per row highlighted in cyan. Each benchmark links to its definition and sources; each model links to its full scorecard.

Frequently asked questions

Which model is best, Kimi K2 Thinking vs Qwen3-30B-A3B? β–Ά

Across the 48 benchmarks compared here, Kimi K2 Thinking leads on the most (27). The right choice still depends on your task β€” check the per-benchmark table above for coding, reasoning and math.

Which of these models is best for coding? β–Ά

Qwen3-30B-A3B has the strongest result on Codeforces Elo among the models compared here. See the code-category rows in the table for the full picture.

How many benchmarks are compared? β–Ά

This comparison covers 48 benchmarks on which at least one of the selected models has a published score.

Where do the benchmark scores come from? β–Ά

Scores are aggregated from official model cards, technical reports and standard public evaluations, and link back to each benchmark's source. They are updated as new results are published.

Explore further

β†’ Kimi K2 Thinking β†’ Qwen3-30B-A3B β†’ AI benchmarks leaderboard β†’ Build a different comparison
Build a different comparison

Models (0 selected)

Benchmarks (auto)

Select at least two models.