Side-by-side comparison · models
Kimi K2.5 vs Kimi K2 Thinking
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
Kimi K2.5
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.
Models
Kimi K2 Thinking
Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.
Checked facts
Line by line
| Measure | Kimi K2.5 | Kimi K2 Thinking |
|---|---|---|
| Input price | $0.532 per million tokens | $0.600 per million tokens |
| Output price | $3.33 per million tokens | $2.50 per million tokens |
| Price on amazon-bedrock | $0.600 per million input tokens | Not checked yet |
| Price on amazon-bedrock | $3.00 per million output tokens | Not checked yet |
| Price on atlas-cloud | $0.200 per million cached input tokens | Not checked yet |
| Price on atlas-cloud | $0.490 per million input tokens | Not checked yet |
| Price on atlas-cloud | $2.50 per million output tokens | Not checked yet |
| Price on novita | $0.095 per million cached input tokens | $0.150 per million cached input tokens |
| Price on novita | $0.570 per million input tokens | $0.600 per million input tokens |
| Price on novita | $2.85 per million output tokens | $2.50 per million output tokens |
| Price on siliconflow | $0.150 per million cached input tokens | Not checked yet |
| Price on siliconflow | $0.500 per million input tokens | Not checked yet |
| Price on siliconflow | $2.50 per million output tokens | Not checked yet |
| Price on venice | $0.209 per million cached input tokens | Not checked yet |
| Price on venice | $0.532 per million input tokens | Not checked yet |
| Price on venice | $3.33 per million output tokens | Not checked yet |
| Price on google-vertex | Not checked yet | $0.600 per million input tokens |
| Price on google-vertex | Not checked yet | $2.50 per million output tokens |
| Latest version | moonshotai/kimi-k2.5-0127 | moonshotai/kimi-k2-thinking-20251106 |
| Time to first token (ms) | 1,780 | 1,750 |
| Time to first token (ms) on amazon-bedrock | 1,140 | Not checked yet |
| Time to first token (ms) on atlas-cloud | 817 | Not checked yet |
| Time to first token (ms) on novita | 1,374 | 1,474 |
| Time to first token (ms) on siliconflow | 1,549 | Not checked yet |
| Time to first token (ms) on venice | 1,011 | Not checked yet |
| Output speed (tokens a second) | 45.4 | 122 |
| Output speed (tokens a second) on amazon-bedrock | 52 | Not checked yet |
| Output speed (tokens a second) on atlas-cloud | 45 | Not checked yet |
| Output speed (tokens a second) on novita | 38 | 44 |
| Output speed (tokens a second) on siliconflow | 22 | Not checked yet |
| Output speed (tokens a second) on venice | 22 | Not checked yet |
| Tokens served on OpenRouter in a day | 3.8B (for 10 Nov 2026) | 338.8M (for 10 Nov 2026) |
| Tool calls on OpenRouter in a day | 46.7K (for 10 Nov 2026) | 4.4K (for 10 Nov 2026) |
| Artificial Analysis Intelligence Index | 23.5 | 22 |
| Artificial Analysis Coding Index | 46.8 | 21 |
| LMArena text rating (Elo) | 1,445 | Not checked yet |
| LMArena web development rating (Elo) | 1,436 | Not checked yet |
| LMArena vision rating (Elo) | 1,269 | Not checked yet |
| Design Arena rating (Elo) | 1,245 | 1,109 |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.