SPECS
| context window | 1,048,576 tokens |
|---|---|
| max output | 943,718 tokens |
| price in / out | $0.003 / $2.4 per 1M tokens |
| weights | open-weights |
openrouter.ai/api/v1/models (as of 2026-10-04)
MODEL / DEEPSEEK-V4-1-FLASH
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
model
textcodevisionreasoning
Kindmodel
Access—
Pricing—
Free tier—
Caps4
PUBLISHED BENCHMARKS
| aaIndex | 39.5 |
|---|
Artificial Analysis via OpenRouter · 2026-09
HUGGING FACE · LIVE
| pipeline | image-text-to-text |
|---|---|
| parameters | 763.2B |
| downloads | 798,422 |
| likes | 4093 |
| last modified | 2026-10-01 |
deepseek-ai/DeepSeek-V4.1-Flash · observed 2026-10-05 00:08:17
MACHINE-READABLE
GET /api/v1/resources/deepseek-v4-1-flashOpen JSON ↗