Compare · ModelsLive · 2 picked · head to head
Kimi K2 Thinking vs Step 3.5 Flash
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Step 3.5 Flash wins on 5/6 benchmarks
Step 3.5 Flash wins 5 of 6 shared benchmarks. Leads in math · knowledge · language.
Category leads
math·Step 3.5 Flashknowledge·Step 3.5 Flashlanguage·Step 3.5 Flashcoding·Step 3.5 Flash
Hype vs Reality
Attention vs performance
Kimi K2 Thinking
#103 by perf·#17 by attention
Step 3.5 Flash
#8 by perf·no signal
Best value
Step 3.5 Flash
11.3x better value than Kimi K2 Thinking
Kimi K2 Thinking
34.0 pts/$
$1.55/M
Step 3.5 Flash
384.5 pts/$
$0.20/M
Vendor risk
Mixed exposure
One or more vendors flagged
Moonshot AI
$18.0B·Tier 1
StepFun
$5.0B·Tier 1
Head to head
6 benchmarks · 2 models
Kimi K2 ThinkingStep 3.5 Flash
OpenCompass · AIME2025
Step 3.5 Flash leads by +1.6
Kimi K2 Thinking
94.1
Step 3.5 Flash
95.7
OpenCompass · GPQA-Diamond
Step 3.5 Flash leads by +1.0
Kimi K2 Thinking
82.7
Step 3.5 Flash
83.7
OpenCompass · HLE
Step 3.5 Flash leads by +0.3
Kimi K2 Thinking
21.3
Step 3.5 Flash
21.6
OpenCompass · IFEval
Step 3.5 Flash leads by +0.8
Kimi K2 Thinking
92.4
Step 3.5 Flash
93.2
OpenCompass · LiveCodeBenchV6
Step 3.5 Flash leads by +6.8
Kimi K2 Thinking
77.1
Step 3.5 Flash
83.9
OpenCompass · MMLU-Pro
Kimi K2 Thinking leads by +0.8
Kimi K2 Thinking
84.3
Step 3.5 Flash
83.5
Full benchmark table
| Benchmark | Kimi K2 Thinking | Step 3.5 Flash |
|---|---|---|
OpenCompass · AIME2025 | 94.1 | 95.7 |
OpenCompass · GPQA-Diamond | 82.7 | 83.7 |
OpenCompass · HLE | 21.3 | 21.6 |
OpenCompass · IFEval | 92.4 | 93.2 |
OpenCompass · LiveCodeBenchV6 | 77.1 | 83.9 |
OpenCompass · MMLU-Pro | 84.3 | 83.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.60 | $2.50 | 262K tokens (~131 books) | $10.75 | |
| $0.10 | $0.30 | 262K tokens (~131 books) | $1.50 |