Updated March 26, 2026· Based on independent benchmark data
Gemini 3 Pro Preview (high) and Kimi K2.5 (Reasoning) are virtually tied on intelligence (48.4 vs 46.8). Kimi K2.5 (Reasoning) is 3.3x cheaper at $0.60/1M tokens vs $2.00/1M. For speed, Gemini 3 Pro Preview (high) wins at 115 tok/s vs 34 tok/s.
| Metric | Gemini 3 Pro Preview (high) | Kimi K2.5 (Reasoning) |
|---|---|---|
| Intelligence Score | 48.4 | 46.8 |
| Coding Score | 46.5 | 39.5 |
| Math Score | 95.7 | N/A |
| Speed (tok/s) | 115 tok/s | 34 tok/s |
| Latency (TTFT) | 36.25s | 1.40s |
| Input Price / 1M tokens | $2.00 | $0.60 |
| Output Price / 1M tokens | $12 | $3.00 |
| Context Window | N/A |
Gemini 3 Pro Preview (high) and Kimi K2.5 (Reasoning) perform similarly on overall intelligence, scoring 48.4 and 46.8 respectively. For coding tasks, Gemini 3 Pro Preview (high) has the edge with a coding score of 46.5 vs 39.5.
Gemini 3 Pro Preview (high) generates output significantly faster at 115 tok/s compared to Kimi K2.5 (Reasoning)'s 34 tok/s, making it 3.4x faster for streaming responses. Time to first token is 1.40s for Kimi K2.5 (Reasoning) vs 36.25s for Gemini 3 Pro Preview (high), which affects perceived responsiveness in interactive applications.
Kimi K2.5 (Reasoning) is more affordable at $0.60/1M input tokens ($3.00/1M output), while Gemini 3 Pro Preview (high) costs $2.00/1M input ($12/1M output). That makes Gemini 3 Pro Preview (high) 3.3x more expensive per token, which can add up significantly at scale. For a typical workload of 100 requests per day at 2,000 tokens each, Gemini 3 Pro Preview (high) would cost approximately $12.00/month vs $3.60/month for Kimi K2.5 (Reasoning) in input costs alone.
Choose Gemini 3 Pro Preview (high) when you need stronger coding performance (46.5), faster output (115 tok/s). Choose Kimi K2.5 (Reasoning) when you need lower cost.
Gemini 3 Pro Preview (high) scores higher on coding benchmarks (46.5 vs 39.5), making it the better choice for programming tasks.
Kimi K2.5 (Reasoning) is cheaper at $0.60/1M input tokens vs $2.00/1M for Gemini 3 Pro Preview (high).
Gemini 3 Pro Preview (high) is faster, producing output at 115 tok/s compared to Kimi K2.5 (Reasoning)'s 34 tok/s.
No, Gemini 3 Pro Preview (high) does not support image input. Neither model supports image input.
Data last synced: March 26, 2026
| N/A |
| Max Output Tokens | N/A | N/A |
| Input Modalities | Text | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
Both models perform similarly on intelligence benchmarks. Choose based on specific needs: pricing, speed, context window, or provider ecosystem.