Skip to main content
Chinese AI Tools
ProductsModelsIntegrationsRankingsLatest changes
TopicsAvailability TrackerUse casesSubmit a toolAccount
ZHSearch

Chinese AI Tools

Independent directory for Chinese AI products. Product availability, pricing and terms can change. Verify before commercial use.

Editorial standardsClaim productUpdate infoGet featuredAdvertise

DeepSeek

DeepSeek

DeepSeek's English API platform for V4-Pro and V4-Flash, with OpenAI-compatible and Anthropic-compatible access.

Globally availableFull English UIPublic APIPaid

Quick answer

The English docs now make DeepSeek's current API shape explicit: V4-Pro/V4-Flash, dual OpenAI/Anthropic compatibility, thinking mode, tool calls, caching and agent-tool integrations.

PricingDocumentation
Pricing
V4-Flash off-peak: $0.007/M cache-hit input, $0.22/M cache-miss input and $0.66/M output; peak rates are double
Availability
Globally available
API
Public API
Last checked
2026-09-07
Pricing detailsUse-case fitSource evidence

DeepSeek is tracked from the official English API docs. The public IDs deepseek-v4-flash and deepseek-v4-pro currently route to V4-Flash-0731 and V4-Pro-0813. Both support 1M context, 384K maximum output, thinking/non-thinking modes, JSON, tools, caching and OpenAI/Anthropic-compatible endpoints. Peak hours are 01:00-04:00 and 06:00-10:00 UTC. V4-Flash costs $0.007/$0.22/$0.66 per million cache-hit input, cache-miss input and output tokens off-peak, or $0.014/$0.44/$1.32 at peak; V4-Pro costs $0.022/$0.66/$1.98 off-peak or $0.044/$1.32/$3.96 at peak.

Topic guide

DeepSeek models, API and agent ecosystem

A source-backed map of DeepSeek V4 models, API pricing, native Harness, third-party agent runtimes and practical integration guides.

Editorial verdict

Best for

Developers evaluating Chinese model APIs for coding agents, reasoning, tool use and OpenAI/Anthropic-compatible migration.

Avoid if

Avoid new integrations that still hard-code deepseek-chat or deepseek-reasoner without a migration plan to V4 model names.

Why it matters

The English docs now make DeepSeek's current API shape explicit: V4-Pro/V4-Flash, dual OpenAI/Anthropic compatibility, thinking mode, tool calls, caching and agent-tool integrations.

Trust: 4/4 sources verified, recently checkedCoverage: 100/100

Pricing

V4-Flash off-peak: $0.007/M cache-hit input, $0.22/M cache-miss input and $0.66/M output; peak rates are double

Payment

Topped-up balance, Granted balance, Platform billing

Commercial use

API commercial use should follow the active platform terms.

Privacy

Treat prompts and logs as vendor-processed data unless enterprise terms say otherwise.

Use-case fit

V4 API migration

Strong

Use deepseek-v4-pro or deepseek-v4-flash directly rather than relying on retired compatibility aliases.

Agent and coding tool backend

Strong

Official docs list integrations for Claude Code, GitHub Copilot, OpenCode, Kilo Code, OpenClaw and other agent tools.

Long-context and cache-aware workloads

Strong

V4 models document 1M context, 384K max output and context caching with cache-hit usage fields.

Global user checklist

RegistrationConfirmedThe English docs link directly to DeepSeek Platform API-key creation.
English UIConfirmedAPI docs, quick start, guides, pricing and changelog are English-facing.
API and docsConfirmedDocs cover OpenAI format, Anthropic format, thinking mode, tool calls, JSON output, FIM, prefix completion and context caching.
International paymentPartialPricing is public per 1M tokens, but top-up/payment availability should be checked inside the platform account.

Pricing and routed snapshots were rechecked on September 7, 2026. The official table still lists deepseek-v4-flash and deepseek-v4-pro as the stable public IDs, routing to V4-Flash-0731 and V4-Pro-0813 with active peak/off-peak billing.

Pros

  • - English API docs are complete and current
  • - OpenAI and Anthropic API compatibility are both documented
  • - V4 models provide 1M context and tool-call support
  • - V4-Flash and V4-Pro pricing distinguishes cache-hit and cache-miss input

Cons

  • - Legacy model names have passed their published deprecation date and should not be used for new integrations
  • - Billing still depends on platform balance and account-specific top-up availability

Decision paths

Choose Qwen for Alibaba Cloud ecosystem fit

Qwen is one option when deployment and cloud billing matter.

Compare with Kimi API for another global-facing model platform

Kimi API is a cataloged reference for English documentation, token billing and agent-oriented models.

Sources

DeepSeek API quick start

Documentation · EN · verified 2026-05-17

Confirms English docs, OpenAI/Anthropic compatibility, V4 model names and legacy alias deprecation date.

DeepSeek models and pricing

Pricing · EN · verified 2026-09-07

Lists V4-Pro-0813 and V4-Flash-0731 with 1M context, 384K maximum output, features, 2,500/500 concurrency limits and active peak/off-peak prices.

DeepSeek agent integrations

Documentation · EN · verified 2026-05-17

Confirms Claude Code and agent-tool integration paths through the Anthropic-compatible endpoint.

DeepSeek change log

Documentation · EN · verified 2026-05-17

Confirms the 2026-04-24 DeepSeek-V4 update and alias deprecation schedule.

Last checked: 2026-09-07

Reviews

No approved reviews yet.

Availability snapshot

Availability
Globally available
English UI
Full English UI
API
Public API
Rating
4.7 (25)

Latest updates

Latest changes
Release · 2026-08-28

Reasonix 1.32 adds remote workspaces and background turns

Reasonix desktop v1.32.0 adds remote SSH workspaces, persistent ordered turns, background turns, conversation navigation, session recovery and Windows fixes. It also moves MCP reliability work onto the official Go SDK and stores additional desktop metadata in SQLite.

Release · 2026-08-27

Hermes Agent 0.20.6 ships a large stable patch rollup

Hermes Agent v0.20.6, tagged v2026.8.27, rolls up roughly 525 pull requests. The release adds consent-gated real-profile browsing, a desktop Browser OS window, SSH remote update and fleet profiles, more than 50 remote MCP servers, web-search and extraction caches, OS keychain encryption and new model integrations.

Release · 2026-08-25

Hermes Agent added with native DeepSeek V4 support

This August 25 catalog update first tracked Nous Research's Hermes Agent at then-stable v0.20.5. Its native DeepSeek provider targeted V4 models, enabled thinking by default, preserved reasoning history across tool calls and migrated retired aliases. Newer Hermes releases are tracked as separate updates.

Release · 2026-08-17

Reasonix 1.26 adds DeepSeek peak/off-peak cost controls

Reasonix 1.26.0 updates its DeepSeek-native coding workflow for the new V4 Flash and V4 Pro peak/off-peak prices. It adds Pro low reasoning effort across OpenAI Chat, Responses and Anthropic modes, unifies cost estimates across CLI, ACP and desktop surfaces, labels peak, off-peak and mixed rates, and automatically migrates configuration to v7.

Submit a review