Models
DeepSeek-V4.1-Flash
Laras Router currently serves one model through a standard OpenAI-compatible endpoint: a fast general-purpose model for everyday AI applications, automated workflows, and high-throughput production workloads.
- Chat completions
- Streaming (SSE)
- Tool calls
- JSON output
- Coding
Specifications
Model
- Model ID
deepseek-flash- Model version
- DeepSeek-V4.1-Flash
- Provider
- DeepSeek
Limits
- Context window
- 1,000,000 tokens
- Maximum output (max_tokens)
- 384,000 tokens
- Default output cap when max_tokens is omitted
- 8,192 tokens
- Per-key limits (default)
- 60 requests / minute · 4 concurrent
Pricing
- Input · cache hit
- $0.010 / 1M tokens
- Input · cache miss
- $0.400 / 1M tokens
- Output
- $1.600 / 1M tokens
- Price by time of day
- Fixed
Capabilities
- Streaming (SSE), tool calls, JSON output
- Supported
- Image input
- Not available
Provider updates: Model availability and pricing may change as DeepSeek releases new versions or revises its service. Legacy model IDs such as
deepseek-v4-flash are not accepted and return 404 model_not_found; use deepseek-flash.