DeepSeek V4 Pro
deepseek · deepseek/deepseek-v4-pro
Open MoE flagship with million-token context for coding and long agent runs
≈667k Chinese characters of context · Step-by-step reasoning · Calls tools / agents · Structured JSON output · Open weights, self-hostable
Read a 200,000-Chinese-character book and write a 2,000-Chinese-character summary: about $0.13
An order-of-magnitude estimate at 1.5 tokens per Chinese character and the vendor's own rate — not a quote. Full method: the glossary on the AI Models page.
Specs
| Context | 1,000,000(≈667k Chinese characters) |
|---|---|
| Max output | 384,000 |
| Structured output | Yes |
| Open weights | Yes |
| Released | 2026-04-24 |
Source: models.dev snapshot 2026-08-29
Who serves it, at what price
| Provider | In /M | Out /M | Cache read /M |
|---|---|---|---|
| DeepSeekfirst-party | $0.435 | $0.87 | $0.003625 |
| Alibaba (China) | $0.435 | $0.87 | $0.003625 |
| Alibaba Token Plan | Billed by plan | Billed by plan | $0 |
| Alibaba Token Plan (China) | Billed by plan | Billed by plan | $0 |
| Auriko | $0.435 | $0.87 | $0.003625 |
| Azure | $1.74 | $3.48 | — |
| Cortecs | $1.73 | $3.46 | $0.432 |
| CrofAI | $0.35 | $0.8 | $0.003 |
| DigitalOcean | $0.87 | $1.74 | $0.174 |
| EmpirioLabs AI | $1.65 | $3.3 | $1.65 |
| FrogBot | $1.74 | $3.48 | $0.14 |
| Charm Hyper | $2.4 | $4.8 | $0.2 |
| Kenari | Billed by plan | Billed by plan | — |
| DevPass (LLM Gateway) | $0.435 | $0.87 | $0.003625 |
| Model Oracle AI | — | — | — |
| Modelis | $0.435 | $0.87 | — |
| Neuralwatt | $1 | $3 | $0.1 |
| Ollama Cloud | — | — | — |
| OpenCode Zen | $1.74 | $3.84 | $0.145 |
| OpenCode Go | $0.66 | $1.98 | $0.022 |
| Requesty | $1.32 | $3.96 | $0.044 |
| routing.run | $0.348 | $0.696 | — |
| UnoRouter | $0.8999 | $1.7999 | — |
| Vancine | $0.66 | $1.98 | $0.022 |
| Venice AI | $1.65 | $3.301 | $0.33 |
| Vivgrid | $0.435 | $0.87 | $0.003625 |
| Volcengine Ark Coding Plan | Billed by plan | Billed by plan | $0 |
Rows tagged “first-party” are the model vendor's own pricing; untagged rows are resale channels and may differ.
When to choose it
This model is on our watch list but has no written assessment yet — we do not write what we cannot support. The specs and prices above still stand.
Leaderboard
Not yet on any leaderboard we track. That does not mean the model performs poorly — it means we have not taken this snapshot for it yet, and we do not publish a score we have not measured.