DeepSeek-V4-Flash vs Phi-3.5-MoE-instruct
DeepSeek vs Microsoft — benchmarks, pricing, and capabilities side by side.
- •DeepSeek-V4-Flash has the higher intelligence index (89.4 vs 49.8)
| DeepSeek-V4-Flash | Phi-3.5-MoE-instruct | |
|---|---|---|
| Intelligence index | 89.4 | 49.8 |
| Developer | DeepSeek | Microsoft |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 1,048,576 tokens | — |
| Input price | $0.10 / 1M | — |
| Output price | $0.20 / 1M | — |
| Speed | 109 tok/s | — |
| Released | April 24, 2026 | August 23, 2024 |
| Parameters | 284B (13B active) | 60000000000 |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
DeepSeek-V4-Flash
Phi-3.5-MoE-instruct
GPQA Diamond
89.4
36.8