Compare models
Put up to 4 models beside each other — token prices, context windows, capabilities and provider, from the same catalogue the model pages read.
| Attribute | MiMo-V2-Flashmimo-v2-flash | Perceptron Mk1.5perceptron-mk1.5 |
|---|---|---|
| Pricing | ||
| Input | $0.09 / 1M | $0.15 / 1M |
| Output | $0.29 / 1M | $1.50 / 1M |
| Cache Write (5m) | $0.09 / 1M | $0.15 / 1M |
| Cache Write (1h) | $0.09 / 1M | $0.15 / 1M |
| Cache Read | $0.09 / 1M | $0.15 / 1M |
| Web Search | $0 / 1M | $0 / 1M |
| Context | ||
| Max context | 262.1K | 36.9K |
| Max output | N/A | N/A |
| Capabilities | ||
| Vision | Yes | No |
| Function Calling | Yes | No |
| JSON Mode | Yes | No |
| Streaming | Yes | Yes |
| Catalogue | ||
| Provider | Xiaomi | Perceptron |
| Category | chat | chat |
| Charge type | Pay As You Go | Pay As You Go |
| Released | — | 2026-09-25 |
| Description | ||
| Summary | MiMo-V2-Flash is an open-source Mixture-of-Experts (MoE) foundation model developed by Xiaomi, featuring 309B total parameters with 15B activated per token and a hybrid attention architecture. It supports a 256K context window and a hybrid thinking mode toggle, enabling flexible trade-offs between speed and reasoning depth. The model excels in reasoning, coding, and agentic workflows, ranking #1 globally among open-source models on benchmarks such as SWE-bench Verified and SWE-bench Multilingual. With performance comparable to leading proprietary models like Claude Sonnet 4.5 at a fraction of the cost, MiMo-V2-Flash is well suited for efficient, high-performance deployments. | Perceptron Mk1.5 chat model that accepts audio input. Audio and text input are billed per token. |