DeepSeek-V3 vs DeepSeek-V4-Pro-0813
Specs, VRAM requirements and download trends — updated 24 October 2026.
deepseek-ai · B · Open weights
Params
1,650.5 B
Context
1,048,576
30d
150 K
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V3 | DeepSeek-V4-Pro-0813 |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 684.5 B | 1,650.5 B |
| Architecture | DeepseekV3ForCausalLM | DeepseekV4ForCausalLM |
| Context length | 163,840 | 1,048,576 |
| Licence | — | mit |
| Commercial use | Unknown | Allowed |
| Languages | — | — |
| Downloads 30d | 1,344,492 | 150,373 |
| Downloads all time | 22.5 M | 380 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-12-25 | 2026-08-13 |
| API input price / 1M tokens | $0.26 | $0.66 |
| API output price / 1M tokens | $1.03 | $1.98 |
| Est. cost per 1M input + 1M output tokens | $1.29 | $2.64 |
| Cached input / 1M tokens | — | $0.022 |
| AA Intelligence Index | — | 36 |
| AA Coding Index | — | 68.8 |
| AA Agentic Index | — | 41.3 |
| Max output tokens | 16,000 | 393,216 |
| API providers | 2 | 21 |
| Tool calling | Yes | Yes |
| Reasoning mode | No | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V3
$1.29
DeepSeek-V4-Pro-0813
$2.64
DeepSeek-V3 is 2.1× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
1.5 M150 K
DeepSeek-V3
DeepSeek-V4-Pro-0813
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
DeepSeek-V3 · fp16 @ 160K ctx
2,811.5 GB / 24 GB
DeepSeek-V4-Pro-0813 · fp16 @ 1024K ctx
32,672.1 GB / 24 GB