DeepSeek V4 Flash 0731 Pricing and Specs Revealed
AccBalanced · x · 2026-08-20
OpenRouter has listed the DeepSeek V4 Flash 0731 model, a sparse mixture-of-experts model.
- Specs: 13B active parameters out of 284B total
- Context: 1,310,720 tokens context, max output 262,144 tokens
- Pricing: $0.0765/M input, $0.153/M output, $0.0153/M Cache Read
- Use Case: Coding, reasoning, and agent workflows
- Release: July 31, 2026 (GA)
OpenRouter also detailed the peak-hour pricing mechanism for the model.
More from Models
- Open Source Models Write Better Than OpenAI/Anthropic, User Claims — JoshPurtell · 2026-08-20
- Google's RT-2 packed 55B parameters, 1000x RT-1's 55M just 7 months later — binarybits · 2026-08-20
- Beating AI Detectors Is a Terrible Proxy for Training Good Writing Models — viemccoy · 2026-08-20
- Gemini 3.7 Flash solves complex network configs — DynamicWebPaige · 2026-08-20
- Qwen3.8-27B Test: Lower KV Cache Quantization Impacts Reasoning Quality — fbms2 · 2026-08-20
- GPT-5.6 learns to look up information online during evals — dejavucoder · 2026-08-20