DeepSeek V4 Flash 0731
by Hetzner · family deepseek-flash · listed Jul 2026RTW
Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding
Current rates USD per 1M tokens
Inputfree
Outputfree
Cache readfree
Cache write—
Against the market
On the record
- Model id
DeepSeek-V4-Flash-0731- Context window
- 512K tokens
- Max output
- 384K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- May 2025
- Released
- 2026-07-31
- Last updated
- 2026-07-31
- Provider docs
- experiments.hetzner.com/docs/inference
Back of the envelope
≈
Also listed at
- Alibaba Token Planfree in · free out
- Alibaba Token Plan (China)free in · free out
- Deep Infra$0.08 in · $0.18 out
- DigitalOcean$0.08 in · $0.252 out
- Eden AI$0.08 in · $0.18 out
- CrofAI$0.12 in · $0.21 out
- Cortecs$0.13 in · $0.28 out
- Inceptron$0.13 in · $0.28 out
- Perplexity Agent$0.13 in · $0.26 out
- Baseten$0.13 in · $0.26 out
- RunInfra$0.13 in · $0.27 out
- Weights & Biases$0.13 in · $0.28 out
- NanoGPT$0.14 in · $0.28 out
- Hugging Face$0.14 in · $0.28 out
- OpenRouter$0.14 in · $0.28 out