Head-to-head showdown: GPT-5.6 Sol ($5.00 in / $30.00 out per 1M) vs Kimi K2.7 Code (OpenRouter) ($0.67 in / $3.40 out per 1M). Kimi K2.7 Code (OpenRouter) is 8.6× cheaper across standard token mixes, with 1.1M vs 262.1K context windows.
gpt-5.6-sol · openai
kimi-k2.7-code · moonshot
Benchmark
| Workload Scenario | GPT-5.6 Sol | Kimi K2.7 Code (OpenRouter) | Price Delta |
|---|---|---|---|
| 1M input tokens (raw text) | $5.00 | $0.67 | +$4.33 |
| 1M output tokens (generation) | $30.00 | $3.40 | +$26.60 |
| 1M tokens · 70% input / 30% output mix | $12.50 | $1.489 | +$11.011 |
| Standard chat turn (4K in / 800 out, 50% cached) | $0.035 | $0.00444 | +$0.0306 |
| Monthly scale (10K requests / day) | $10,500.00 | $1,332.00 | +$9,168.00 |
Negative difference = GPT-5.6 Sol is cheaper. Positive = Kimi K2.7 Code (OpenRouter) is cheaper.
Scaling Curve
Total cost of a token volume at 70/30 input/output split (uncached). Log-log scale.
Verdict
Kimi K2.7 Code (OpenRouter) is cheaper on both input and output rates, so it costs less at every input/output mix. Price alone still isn't the whole decision: capability, latency and context limits (GPT-5.6 Sol: 1.1M, Kimi K2.7 Code (OpenRouter): 262.1K) may justify the premium for your task.
FAQ
On input, Kimi K2.7 Code (OpenRouter) is cheaper ($0.67/M vs $5.00/M). On output, Kimi K2.7 Code (OpenRouter) is cheaper ($3.40/M vs $30.00/M). The same model is cheaper on both sides, so it wins at every mix.
A chat-style request (4,000 input + 800 output tokens, 50% cached) costs $0.035 on GPT-5.6 Sol and $0.00444 on Kimi K2.7 Code (OpenRouter) — Kimi K2.7 Code (OpenRouter) is 7.9× more expensive for that workload.
GPT-5.6 Sol supports 1,050,000 tokens (128K max output); Kimi K2.7 Code (OpenRouter) supports 262,144 (235.9K max output). GPT-5.6 Sol fits 4.0× more context, which matters for long documents and agents.