DeepSeek V4 Pro
by Charm Hyper · family deepseek-thinking · listed Jul 2026RTW
Open MoE flagship with million-token context for coding and long agent runs
Current rates USD per 1M tokens
Input$2.40
Output$4.80
Cache read$0.20
Cache write—
Against the market
On the record
- Model id
deepseek-v4-pro- Context window
- 1M tokens
- Max output
- 384K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- May 2025
- Released
- 2026-07-06
- Last updated
- 2026-07-22
- Provider docs
- hyper.charm.land
Back of the envelope
≈
Also listed at
- Kenarifree in · free out
- Alibaba Token Planfree in · free out
- Alibaba Token Plan (China)free in · free out
- routing.run$0.348 in · $0.696 out
- CrofAI$0.35 in · $0.80 out
- EBCloud$0.429 in · $0.857 out
- OpenCode Go$0.435 in · $0.87 out
- Nvidia$0.435 in · $0.87 out
- Vivgrid$0.435 in · $0.87 out
- LLM Gateway$0.435 in · $0.87 out
- Pioneer$0.435 in · $0.87 out
- ZenMux$0.435 in · $0.87 out
- Auriko$0.435 in · $0.87 out
- DeepSeek$0.435 in · $0.87 out
- Kilo Gateway$0.435 in · $0.87 out