DeepSeek First-PartyOpen-WeightVerified 2026-11-12
DeepSeek Flash (V4.1-Flash-1017) API Pricing & Cost Rules
API Model ID: deepseek-flash · Context Window: 1,000,000 tokens · Max Output: 393,216 tokens
Context Tiers & Token Rate Card (USD per 1M Tokens)
| Tier | Std Input | Std Output | Cache Read | Write (Short) | Write (Long) |
|---|---|---|---|---|---|
| Standard (<= 1M) | $0.3 | $1.2 | $0.006 | $0.3 | N/A |
Prompt Caching Rules
Supported: Yes (automatic)
Minimum Cacheable Prefix: None / Automatic
Hourly Cache Storage: $0
Automatic disk-based KV caching with 98% discount (0.02x): $0.006/1M Peak, $0.003/1M Off-Peak. Zero cache write surcharge.
Batch API & Hosting Notes
Batch Supported: No
Caching in Batch: Disabled in Batch mode
No Batch API exists on DeepSeek (confirmed via official sitemap.xml). Schedule jobs in Off-Peak UTC windows for 50% off.
2,500 concurrency limit; supports OpenAI and Anthropic API formats.