Kimi K2-Thinking-0905 vs Kimi K3: Specs & Benchmark Comparison
| Characteristic | Kimi K2-Thinking-0905 | Kimi K3 |
|---|---|---|
| Company | Moonshot AI | Moonshot AI |
| Release Date | September 4, 2025 | July 16, 2026 |
| Parameters | 1.0T | 2.8T |
| Multimodal | No | Yes |
| Context (input) | 262K | 1.0M |
| Context (output) | 66K | 1.0M |
| Input Price / 1M | $0.60 | $3.00 |
| Output Price / 1M | $2.40 | $15.00 |
| Average Score | 0.9 | 1.0 |
| Benchmarks | ||
| GPQA | 0.8 | 0.9 |
Visual Benchmark Comparison
Kimi K2-Thinking-0905
Kimi K3
GPQA0.8 vs 0.9
0.8
0.9
Verdict
Kimi K3 leads in 2 out of 4 comparison categories.
Overall Performance
Both models show comparable average scores: Kimi K2-Thinking-0905 — 0.9, Kimi K3 — 1.0.
API Cost
Kimi K2-Thinking-0905 is 6.0x cheaper: input $0.60/1M vs $3.00/1M tokens.
Context Window
Kimi K3 supports a larger context: 1M vs 262K tokens.
Recency
Kimi K3 is newer: released 7/16/2026 vs 9/4/2025.
More About These Models
Related Comparisons
Frequently Asked Questions
Which is better for coding — Kimi K2-Thinking-0905 or Kimi K3?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — Kimi K2-Thinking-0905 or Kimi K3?
Kimi K2-Thinking-0905 is cheaper for input: $0.60 per 1M tokens vs $3.00.
Which has a larger context window — Kimi K2-Thinking-0905 or Kimi K3?
Kimi K3 supports a larger context: 1,000,000 tokens vs 262,144.
The Kimi K2-Thinking-0905 and Kimi K3 comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Kimi K2-Thinking-0905 or Kimi K3 page. See also the complete list of AI model comparisons.