DeepSeek API Pricing (August 2026)

DeepSeek exposes an OpenAI-compatible base URL at https://api.deepseek.com. Create a key in the DeepSeek platform, keep it on the server, and send it as a Bearer token. This request uses the current Flash model name rather than the retired deepseek-chat alias:

curl https://api.deepseek.com/chat/completions \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $DEEPSEEK_API_KEY" \
  -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Reply with OK"}]}'

A successful response includes token usage. Do not place the API key in browser code, logs, or a public repository. The LLM API pricing calculator can then turn measured input, output, and cache-hit tokens into a monthly estimate.

DeepSeek's current pricing table lists cache-hit input, cache-miss input, and output rates for V4 Flash and V4 Pro. A separate notice says a broader API price increase is planned, but DeepSeek has not published replacement rates or an effective date.

So the table, structured price data, and calculator keep using the published values. The calculator does not apply an unpublished increase, and we will not invent a multiplier to fill the gap.

DeepSeek prices input tokens in two lanes: cache miss (the standard rate) and cache hit, billed when the service can reuse a prompt prefix. A V4 Flash cache hit costs $0.0028 per million tokens; V4 Pro costs $0.003625. Those are explicit provider rates, not a percentage inferred from the standard input price.

Agents, RAG pipelines, and chat applications can benefit when they resend the same system prompt or document context. But the table does not promise a hit rate: a workload with mostly new prefixes pays the cache-miss rate. Enter your measured cached share in the estimator above. DeepSeek lists no separate Batch API tier.

The response exposes prompt_cache_hit_tokens and prompt_cache_miss_tokens in its usage object. Log those two fields and calculate the real hit share before budgeting. DeepSeek says unused cache entries are usually cleared after hours to days, so a one-time warm-up does not establish a permanent discount.

DeepSeek currently lists account-level concurrency limits of 500 for V4 Pro and 2,500 for V4 Flash. A request occupies one slot until its response completes. Requests beyond the account limit receive HTTP 429, regardless of how many API keys the account uses.

Production clients should cap concurrency, retry 429 responses with jittered backoff, and keep a fallback path for availability-sensitive work. DeepSeek also documents empty-line or SSE keep-alive traffic while a request waits; custom HTTP parsers must ignore those keep-alives without treating them as the JSON result.

V4 Pro ($0.435/$0.87 regular) is the flagship, with a 1M-token context window and separate reasoning-effort rows in the current Chinese model rankings. Use it for coding, reasoning, and agentic work where quality clears your workload's tests. V4 Flash ($0.14/$0.28 regular) is the volume tier — classification, extraction, summarization, and high-throughput pipelines where per-request cost dominates.

Older V3.2 and R1rows remain in the registry for historical comparison. DeepSeek retired the legacy deepseek-chat and deepseek-reasoner aliases after July 24, 2026 at 15:59 UTC. New requests must use deepseek-v4-flash or deepseek-v4-pro. DeepSeek published the April V4 Preview checkpoints as open weights, so those can be self-hosted; it had not published weights for the 0731 post-training update when we checked on July 31.

The current prices and the legacy rows answer different questions. Use the V4 rows for a new API integration; use the older rows only to reconstruct a historical bill or migration. DeepSeek changes the model behind an alias, so a familiar model ID does not prove that the underlying version stayed fixed.

The price table above links every row to DeepSeek's current or archived first-party evidence and gives the effective version date. BenchLM does not treat historical alias pricing as a currently callable SKU.

The gap is not subtle. V4 Pro at $0.435/$0.87 costs about 14x less than GPT-5.6 Terra ($2/$12) on output and more than 28x less than GPT-5.6 Sol or Claude Opus 4.8 ($30 and $25 output respectively). V4 Flash at $0.14/$0.28 undercuts even OpenAI's nano tier on output. The honest caveat: the Western flagships still score higher on BenchLM's overall leaderboard, so the decision is quality-per-dollar, not price alone.

Compare the full tables: OpenAI API pricing and Claude API pricing, or use the current product-level DeepSeek vs Claude comparison. The Opus 5 vs V4 Pro model page keeps the benchmark-by-benchmark evidence separate.