Kimi k1.5 vs Phi-4
Moonshot AI vs Microsoft · Prices checked 12 August 2026
There is a 35.7× price gap here. Phi-4 blends to $0.133 per million tokens; Kimi k1.5 to $4.75. A single 100K-input, 10K-output job costs $0.0098 on Phi-4 against $0.350 on Kimi k1.5.
Capability favours Kimi k1.5, ahead on 4 of 4 suites they both publish. The widest gap is on MATH, where Kimi k1.5 leads by 16 points. Kimi k1.5 reads more in one request: 128,000 tokens of context against 16,000.
Pricing and capacity
| Metric | Kimi k1.5 | Phi-4 | Difference |
|---|---|---|---|
| Input / 1M tokens | $2.50 | $0.070 | 35.7× |
| Output / 1M tokens | $10.00 | $0.280 | 35.7× |
| Blended · 70/30calculated | $4.75 | $0.133 | 35.7× |
| Blended · 50/50calculated | $6.25 | $0.175 | 35.7× |
| Blended · 80/20calculated | $4.00 | $0.112 | 35.7× |
| Blended · 20/80calculated | $8.50 | $0.238 | 35.7× |
| Job cost · 100K + 10Kcalculated | $0.350 | $0.0098 | 35.7× |
| Job cost · 1M + 100Kcalculated | $3.50 | $0.098 | 35.7× |
| Context window | 128,000 | 16,000 | 8.0× |
Input and output rates are as published by each provider. Blended and job-cost rows are calculated comparison metrics, not published prices — how they are derived.
Benchmarks
| Benchmark | Kimi k1.5 | Phi-4 | Leader |
|---|---|---|---|
| MMLU | 88% | 84% | Kimi k1.5 |
| GPQA Diamond | 70% | 56% | Kimi k1.5 |
| MATH | 96% | 80% | Kimi k1.5 |
| HumanEval | 92% | 82% | Kimi k1.5 |
Only suites both models report are listed, so a missing score is never shown as a loss. Figures are provider-published or from public leaderboards and are not independently re-run.
What each model is for
Kimi k1.5
Moonshot AI · Reasoning
Moonshot's reasoning Kimi. RL-trained chain-of-thought competitive with o1 on math benchmarks; long-context strength is the Kimi signature.
- Reasoning + long context
- Strong math
- Visible thinking
- Aggressive Chinese pricing
Frequently asked questions
Which is cheaper, Kimi k1.5 or Phi-4?
There is a 35.7× price gap here. Phi-4 blends to $0.133 per million tokens; Kimi k1.5 to $4.75. A single 100K-input, 10K-output job costs $0.0098 on Phi-4 against $0.350 on Kimi k1.5. Blended cost weights input at 70% and output at 30%; price your own token split for an exact answer, since the cheaper model on a blended basis is not always cheaper for your workload.
Which performs better, Kimi k1.5 or Phi-4?
Capability favours Kimi k1.5, ahead on 4 of 4 suites they both publish. The widest gap is on MATH, where Kimi k1.5 leads by 16 points. These are the figures each provider or public leaderboard publishes, not results we re-run, and a benchmark average can hide the specific behaviour your workload depends on.
Which has the larger context window, Kimi k1.5 or Phi-4?
Kimi k1.5 reads more in one request: 128,000 tokens of context against 16,000. The context window caps everything in a single request: prompt, retrieved documents, conversation history and the model’s own output all share it.
Can I compare Kimi k1.5 and Phi-4 on my own numbers?
Yes. Open both in the side-by-side comparison tool to see every field at once, or use the token calculator to price your actual prompt and monthly request volume against each model. Both are free and need no sign-up.
Price it for your own workload
Related matchups
Terms used on this page
Prices and scores are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.