DeepSeek: DeepSeek V4 Pro 0423
DeepSeekOpen weights
DeepSeek: DeepSeek V4 Pro 0423 by DeepSeek costs $0.96 per 1M input tokens and $1.91 per 1M output tokens, with a 1M-token context window.
Pricing
Input (per 1M tokens)
$0.96
Output (per 1M tokens)
$1.91
Cached input (per 1M)
$0.08
| Provider | Input / 1M | Output / 1M |
|---|---|---|
| openrouter | $0.96 | $1.91 |
| DeepSeek | $0.66 | $1.98 |
Specifications
- Provider
- DeepSeek
- Context window
- 1M tokens
- Parameters
- 1.6T
- Released
- Apr 2026
- Open weights
- Yes
- Frontier model
- No
- Hugging Face
- deepseek-ai/DeepSeek-V4-Pro
Benchmarks
- GPQA Diamond
- 91.7
- LiveBench
- 71.6
- LMArena WebDev Elo
- 1582.2
Speed & latency
Time to first token
2.64s
Output speed
187 tok/s
Fastest host: azure (via vercel-ai-gateway).
Compare DeepSeek: DeepSeek V4 Pro 0423 with…
DeepSeek: DeepSeek V4 Pro 0423 vs Qwen3 VL 235B A22B Instruct$1.90/1MDeepSeek: DeepSeek V4 Pro 0423 vs Qwen3.6 Plus$1.95/1MDeepSeek: DeepSeek V4 Pro 0423 vs Qwen3.5 35B-A3B$2.00/1MDeepSeek: DeepSeek V4 Pro 0423 vs GPT-5 Mini$2.00/1MDeepSeek: DeepSeek V4 Pro 0423 vs GLM 4.6$2.00/1MDeepSeek: DeepSeek V4 Pro 0423 vs GPT-5.1-Codex-Mini$2.00/1M
FAQ
How much does DeepSeek: DeepSeek V4 Pro 0423 cost?
DeepSeek: DeepSeek V4 Pro 0423 is priced at $0.96 per 1M input tokens and $1.91 per 1M output tokens. Output tokens cost more than input.
What is DeepSeek: DeepSeek V4 Pro 0423's context window?
DeepSeek: DeepSeek V4 Pro 0423 supports a context window of 1M tokens — the maximum prompt plus response length it can handle in a single request.
Is DeepSeek: DeepSeek V4 Pro 0423 open weight?
Yes — DeepSeek: DeepSeek V4 Pro 0423 has openly available weights.
Data as of 2026-10-16. Source: Pricing via OpenRouter API, openrouter.ai. Pricing is per 1M tokens (USD); confirm with the provider before production use.