Official docsRetrieved 2026-05-20Fresh
DeepSeek — Chat completion API referenceModel intelligence
DeepSeek V4 Pro
DeepSeek's current generation reasoning model. Listed by DeepSeek alongside deepseek-v4-flash on the API Models & Pricing page; the older deepseek-chat and deepseek-reasoner models are documented as scheduled for deprecation on 2026/07/24.
- Snapshot date
- Data not yet verified.
- Context window
- 1,000,000 tokenssrc
- Max output
- Data not yet verified.
- Modality
- Data not yet verified.
Data freshness: last checked 8 d ago · updated 2026-05-21
Verification
4 of 10 tracked fields verified against a primary source.
VerifiedActive
- API identifiersverified
- Release / snapshot datenot yet verified
- Context windowverified
- Max output tokensnot yet verified
- Modalitynot yet verified
- Knowledge cutoffnot yet verified
- Featuresnot yet verified
- Lifecycleverified
- Pricingverified
- Benchmarksnot yet verified
Data freshness: last checked 8 d ago · updated 2026-05-21
Verification checklist
4 of 10 tracked fields verified against a primary source
Each row reflects the actual state on the record — no checkmark is asserted without isVerified() returning true.
- Verified
API identifiers (canonical / alias)
- Verified
Context window
- Missing
Max output tokens
- Missing
Modality channels
- Missing
Knowledge cutoff
- Missing
Features (extended thinking / tool use / vision …)
- Verified
Lifecycle (active / deprecated / retired)
- Missing
Release date
- Missing
Snapshot date
- Verified
Pricing rows
3 pricing rows sourced
Data gaps
Fields not yet verified
These fields are intentionally null until verified against a primary source. They are not estimated.
Max output tokens
Provider docs do not publish a max-output number separately from the context window for this model, or the field has not yet been retrieved.
Modality channels
Provider docs describe the model without enumerating which input/output channels are supported, or the field has not yet been retrieved.
Knowledge cutoff
No reliable cutoff date published in provider docs for this model.
Benchmark scores
No third-party benchmark scores have been verified for any model on this site — see the benchmark-limitations guide for why.
Infrastructure (regions / latency / uptime)
Infrastructure fields are intentionally null until a durable measurement source is wired. See the inference-infrastructure guide.
How a value moves from this list to a verified row — see source verification methodology.
Reference
API identifiers
Pinned snapshot IDs as listed by the provider.
- Canonical
- deepseek-v4-pro
API pricing
First-party API pricing references
Per-unit rates published by DeepSeek for its own first-party API. Each row is a source-backed reference, not a live quote. Where DeepSeek does not run a paid first-party API for this model, the table renders the unverified-data label and any hosted-provider rates appear below.
Pricing is volatile and may change after retrieval. Treat prices as source-backed references, not live quotes.
| Unit | Price (USD) |
|---|---|
| 1M input tokens | $1.74src |
| 1M cache read tokens | $0.01src |
| 1M output tokens | $3.48src |
Hosted provider pricing references
Third-party platforms that host this model and bill at their own rate. Hosted pricing is billing-provider-specific and may change. It is not model-creator pricing unless the creator is also the billing provider. See methodology.
Hosted by Together AI
deepseek-ai/DeepSeek-V4-ProVolatility: highFreshness: Freshreview every 14d
| Unit | Price (USD) |
|---|---|
| 1M input tokens | $2.10src |
| 1M output tokens | $4.40src |
| 1M cache read tokens | $0.20src |
DeepSeek V4 Pro's Together rate is set by Together AI, not DeepSeek. The two billing providers price independently — this row is a source-backed reference, not a comparison datapoint.
Capability
Benchmarks
Benchmark scores are only published once verified against a primary source. WebmasterID Models does not republish provider-reported scores without an independent reference.
Infrastructure
Inference infrastructure
Regions, latency, and uptime where independently verified.
Regions
Data not yet verified.
Avg latency (ms)
Data not yet verified.
Uptime (%)
Data not yet verified.
Compare
Related comparisons
Continue
Related models
Other models in the catalogue — same provider first, then the rest of the graph.
Entity graph
Provider context
- Provider page: /providers/deepseek
- Pricing rows for this provider: /pricing?provider=deepseek
- Sources for this provider: /sources?provider=deepseek
Documentation example
Calling DeepSeek V4 Pro via the DeepSeek chat-completions API
Shape of a minimal chat-completions request using the verified canonical model ID. The model parameter value, endpoint, and minimum payload are taken directly from DeepSeek's API reference.
curl
curl https://api.deepseek.com/chat/completions \
-H "Authorization: Bearer $DEEPSEEK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-pro",
"messages": [
{ "role": "user", "content": "Hello, DeepSeek V4 Pro." }
]
}'Shown for reference — not a performance recommendation. Refer to the provider's documentation for current request shape, headers, and limits.
Sources
- Official pricingRetrieved 2026-05-20FreshDeepSeek — Models & Pricing
DeepSeek API Models & Pricing page — current model IDs (deepseek-v4-flash, deepseek-v4-pro), context window, input cache-miss / cache-hit / output rates, time-boxed discount window.
- Official docsRetrieved 2026-05-20FreshDeepSeek — Chat completion API reference
DeepSeek chat completion API reference — endpoint, model parameter values (deepseek-v4-flash, deepseek-v4-pro), minimal request body.
- Official docsRetrieved 2026-05-20FreshDeepSeek — API docs root
DeepSeek API docs root — active vs deprecated model status (deepseek-chat and deepseek-reasoner marked for deprecation 2026/07/24).
Verification queue
Next review for this entity targets re-checking pricing against official provider documentation and confirming lifecycle state. Cadence per field type lives in /docs.