DeepSeek API Pricing Calculator & Complete Cost Guide
Calculate DeepSeek V4 (Flash & Pro) API costs per token and per month. Up to 99% cheaper than GPT-5.5 with automatic context caching.
Pricing TLDR
- • 5 million free tokens for new users (no credit card required)
- • V4 Flash: $0.14 input • $0.28 output per million tokens (up to 99% cheaper than GPT-5.5)
- • V4 Pro: $1.74 input • $3.48 output for flagship reasoning • 1M context window • 384K max output
DeepSeek API Cost Calculator - Monthly Pricing
Calculate by
Input Tokens
Output Tokens
API Calls / Month
Quick Examples:
Context Caching:
Cache Hit Rate:
50%
(V4 Flash: $0.0028 cache hit vs $0.14 miss; V4 Pro: $0.0145 vs $1.74)
DeepSeek automatically caches context. When requests share the same prefix, cached segments are reused.
DeepSeek V4 Pro (deepseek-v4-pro)
Context
384K output
Quality
Per 1M Tokens
In: $0.88
Out: $3.48
(50% cache hits)
Monthly Cost
DeepSeek V4 Flash (deepseek-v4-flash)
Context
384K output
Quality
Per 1M Tokens
In: $0.07
Out: $0.28
(50% cache hits)
Monthly Cost
DeepSeek V4 Pro (deepseek-v4-pro)
Context
384K output
Quality
Per 1M Tokens
In: $0.88
Out: $3.48
(50% cache hits)
Monthly Cost
DeepSeek V4 Flash (deepseek-v4-flash)
Context
384K output
Quality
Per 1M Tokens
In: $0.07
Out: $0.28
(50% cache hits)
Monthly Cost
Burning through DeepSeek API credits?
Track your DeepSeek API spending in real-time.
Privacy-first desktop app. No sign-up required.
About DeepSeek API
What is DeepSeek API?
The DeepSeek API provides programmatic access to DeepSeek's V4 family — Flash (cost-effective) and Pro (flagship) — each available in thinking and non-thinking modes. V4 unifies the context window at 1M tokens and max output at 384K tokens across all modes, with automatic context caching to reduce costs for repeated prompts. The legacy `deepseek-chat` and `deepseek-reasoner` API names continue to work as compatibility aliases for V4 Flash modes.
- Extremely Cost-Effective: DeepSeek V4 Flash is one of the most affordable high-capability APIs available. At $0.14/$0.28 per million tokens (cache miss / output), it's up to 99% cheaper than GPT-5.5 ($5/$30). V4 Pro at $1.74/$3.48 still undercuts Claude Sonnet 4.6 ($3/$15) for near-flagship reasoning quality.
- Automatic Context Caching: Context caching is enabled by default across both tiers. When requests share the same prefix as recent ones, cached segments are retrieved from disk automatically. Cache hits cost just $0.0028/M tokens on V4 Flash (98% cheaper than cache miss) and $0.0145/M on V4 Pro.
- Thinking & Non-Thinking Modes: Both V4 Flash and V4 Pro offer thinking mode (visible Chain-of-Thought, best for math/logic/code) and non-thinking mode (faster, best for general tasks). V4 unifies max output at 384K tokens across both modes — no more 8K vs 64K split.
When to Use DeepSeek API
DeepSeek is ideal for cost-sensitive applications that need good AI capability without frontier pricing. Use V4 Flash for high-volume production workloads and V4 Pro for hard reasoning tasks; within each tier, enable thinking mode for Chain-of-Thought reasoning or use non-thinking mode for general-purpose workloads.
Ideal for
- Cost-sensitive production applications
- High-volume batch processing
- Math, logic, and coding tasks (use thinking mode)
- General chatbots and content generation (use non-thinking mode)
- Applications with repetitive prompts (benefits from caching)
Not ideal for
- Applications requiring maximum frontier capability
- Use cases needing specific tool integrations not offered
- Workloads requiring strict SLAs or guaranteed uptime
- Tasks requiring guaranteed deterministic outputs
DeepSeek API Pricing Breakdown
Free Tier
New users receive 5 million free tokens upon registration with no credit card required. These credits are automatically applied to your usage and work across all models.
- Sign up at platform.deepseek.com - no credit card required
- Receive 5 million free tokens instantly
- Credits work across all models (V4 Flash and V4 Pro)
- Additional credits can be purchased as needed
Key Features
Automatic Context Caching
All requests automatically benefit from context caching. When prompts share the same prefix, cached content is reused. On V4 Flash, cache hits cost $0.0028/M vs $0.14/M for cache misses — a 98% savings. (The cache-hit rate was reduced to 1/10 of launch price on 2026-04-26.)
V4 Two-Tier Pricing
DeepSeek V4 ships in two tiers. V4 Flash: $0.0028 cache hit, $0.14 cache miss, $0.28 output per 1M — for cost-effective production workloads. V4 Pro: $0.0145 cache hit, $1.74 cache miss, $3.48 output per 1M — for flagship reasoning. Both tiers offer thinking and non-thinking modes.
1M Context, 384K Max Output
V4 ships with a 1M-token context window and unified 384K max output tokens across both thinking and non-thinking modes — a major expansion from V3.2's 128K context / split 8K-64K outputs.
OpenAI- and Anthropic-Compatible API
DeepSeek API speaks two formats — OpenAI-compatible at the default base URL and Anthropic-compatible at `https://api.deepseek.com/anthropic`. Migrate from either ecosystem by changing your base URL and API key.
DeepSeek Model Comparison
Quality (non-thinking / thinking)
DeepSeek V4 Flash
DeepSeek V4 Pro
Cache hit / miss / output (per 1M)
DeepSeek V4 Flash
DeepSeek V4 Pro
Context Window
DeepSeek V4 Flash
DeepSeek V4 Pro
Max Output
DeepSeek V4 Flash
DeepSeek V4 Pro
Chain-of-Thought
DeepSeek V4 Flash
DeepSeek V4 Pro
Best For
DeepSeek V4 Flash
DeepSeek V4 Pro
Legacy aliases
DeepSeek V4 Flash
DeepSeek V4 Pro
Note: V4 Pro is currently running a 75% launch discount until 2026-05-31 15:59 UTC. The prices above show the post-promo steady-state rates so the calculator stays accurate after the promo window. The `deepseek-chat` and `deepseek-reasoner` model names still work as compatibility aliases routing to V4 Flash modes, but are scheduled for deprecation on 2026-07-24 — new integrations should use the explicit `deepseek-v4-flash` and `deepseek-v4-pro` model names.
DeepSeek API Monthly Cost Estimates
Light Use
$1-5/mo
• Personal projects
• <1K requests/day
• Either mode works
Medium Use
$5-25/mo
• Small apps
• 1-5K requests/day
• Non-thinking for general, thinking for reasoning
Heavy Use
$25-125/mo
• Production apps
• 5-20K requests/day
• Optimize cache hit rate
Enterprise
$125+/mo
• Large scale
• 20K+ requests/day
• High cache utilization
7 DeepSeek API Cost Optimization Tips
Maximize Cache Hit Rate
Structure prompts with consistent prefixes (system prompts, instructions) to maximize cache hits. On V4 Flash, cache hits cost $0.0028/M vs $0.14/M for misses — a 98% savings. Aim for 70%+ cache hit rates in production.
Start with V4 Flash, Upgrade to Pro Selectively
V4 Flash (quality 61 non-thinking / 77 thinking) handles classification, summarization, general chat, and most coding at 12x lower cost than V4 Pro. Reserve V4 Pro (quality 65 / 86) for hard reasoning, complex math, and flagship-quality code generation.
Use Non-Thinking Mode for Simple Tasks
Within a tier, non-thinking mode is faster and well-suited for classification, summarization, and general queries. Reserve thinking mode for complex math, logic, and code tasks where Chain-of-Thought pays off in quality.
Batch Similar Requests
Group requests with similar prompts together to benefit from context caching. The system automatically caches and retrieves shared prefixes, reducing costs on subsequent requests.
Optimize Output Length
Both V4 modes support up to 384K output tokens, but costs accumulate per token. Set appropriate max_tokens limits for your use case — most tasks finish well under 10K output tokens.
Compare with Competitors
V4 Flash at $0.14/$0.28 is up to 99% cheaper than GPT-5.5 ($5/$30) and Claude Sonnet 4.6 ($3/$15). V4 Pro at $1.74/$3.48 is still meaningfully cheaper than Claude Opus 4.7 ($5/$25) for near-flagship reasoning quality (86 vs Opus's 95).
Monitor Token Usage
Track your cache hit rates and token consumption via the DeepSeek platform. Understanding your caching patterns helps optimize prompt design and reduce costs further.
DeepSeek Model Selection Guide
Use Case
Customer Support Chat
Recommended Model
V4 Flash
Non-thinking mode
Monthly Cost (Est.)
~$1-4
Why This Model?
Fast, cheapest for general queries
Use Case
Code Generation
Recommended Model
V4 Flash
Thinking mode
Monthly Cost (Est.)
~$3-12
Why This Model?
Quality 77 with Chain-of-Thought at Flash pricing
Use Case
Hard Math & Logic
Recommended Model
V4 Pro
Thinking mode
Monthly Cost (Est.)
~$20-80
Why This Model?
Quality 86 for the hardest reasoning tasks
Use Case
Content Writing
Recommended Model
V4 Flash
Non-thinking mode
Monthly Cost (Est.)
~$2-8
Why This Model?
Cheap for general content, 384K output ceiling
Use Case
Data Extraction
Recommended Model
V4 Flash
Non-thinking mode
Monthly Cost (Est.)
~$1-4
Why This Model?
JSON output support, tool calling, lowest cost
Use Case
High-Volume Batch
Recommended Model
V4 Flash
With high cache rate
Monthly Cost (Est.)
~$2-20
Why This Model?
Maximize cache hits at $0.0028/M for the lowest cost
Start Tracking Your DeepSeek API Spending
Monitor your DeepSeek V4 credit balance and per-tier usage from your menubar. Alerts before you run out.
Privacy-first desktop app. 7-day free trial, no credit card required.
DeepSeek API Pricing FAQ
Common questions about DeepSeek API costs, billing, and optimization
