LLM Comparison

Kimi K3 vs MiniMax M2.5: benchmark scores, pricing & comparison.

Side-by-side Kimi K3 vs MiniMax M2.5 comparison across SWE-bench, GPQA, HLE, Terminal-Bench, coding agent scores, token pricing, context window, and AskClash RWT. Green marks the winner on each benchmark.

Open live leaderboardKimi K3 model pageMiniMax M2.5 model page
Rank #5 vs #39AskClash overall scores 76.4 vs 36.0.
Pricing $3.00/$15.0 vs $0.30/$1.20Input and output token prices per 1M tokens when published.
Open Weight vs ProprietaryMoonshot AI vs MiniMax.

Kimi K3 vs MiniMax M2.5 benchmark comparison

Green cells highlight the winning model for each metric. Scores are cached from the AskClash LLM leaderboard snapshot.

MetricKimi K3MiniMax M2.5
Overall Score76.436.0
Leaderboard Rank#5#39
RWT8.0—
HLE43.519.1
GPQA93.584.8
IFEval—71.6
SWE-bench—80.2
SWE-Pro—55.4
Terminal-Bench88.3—
DeepSWE68.5—
MCP Atlas84.2—
Finance Agent54.4—
CharXiv84.8—
MMMU-Pro81.6—
Tau2—95.3
Input Price (per 1M tokens)$3.00$0.30
Output Price (per 1M tokens)$15.0$1.20
Context Window1M128K
Benchmarks Published118

Kimi K3 vs MiniMax M2.5 head-to-head charts

Kimi K3 leads 3 and MiniMax M2.5 leads 0 of 3 shared benchmarks. MiniMax M2.5 is cheaper on a defined 1M-input / 200K-output workload. Charts show only benchmarks both models publish.

Kimi K3MiniMax M2.5
Overall
76.4Kimi K3
36.0MiniMax M2.5
HLE
43.5Kimi K3
19.1MiniMax M2.5
GPQA
93.5Kimi K3
84.8MiniMax M2.5
Kimi K3
Input$3.00
Output$15.0
Workload$6.00
Context1M
MiniMax M2.5
Input$0.30
Output$1.20
Workload$0.54
Context128K

Workload = published cost of 1M input + 200K output tokens. Open the live leaderboard for interactive compare charts.

More Kimi K3 and MiniMax M2.5 comparisons

Explore how Kimi K3 and MiniMax M2.5 stack up against other top-ranked LLMs.

Claude Opus 5 vs Kimi K3Claude Opus 5 vs MiniMax M2.5Claude Fable 5 vs Kimi K3Claude Fable 5 vs MiniMax M2.5GPT-5.6 Sol vs Kimi K3GPT-5.6 Sol vs MiniMax M2.5Claude Opus 4.8 vs Kimi K3Claude Opus 4.8 vs MiniMax M2.5Kimi K3 vs Claude Opus 4.7Claude Opus 4.7 vs MiniMax M2.5

How to read this comparison

Benchmark scores

Higher is better for all benchmark scores (SWE-bench, GPQA, HLE, Terminal-Bench, etc.). Green marks the model with the higher score.

Token pricing

Lower is better for input and output prices. Green marks the cheaper model per 1M tokens.

Coverage matters

Models with fewer disclosed benchmark cells may have inflated percentile scores. Check the benchmark cell count for context.

This comparison page is generated from the AskClash LLM leaderboard cache. Open the live leaderboard for real-time scores and interactive filtering.