API PRICES

DeepSeek · deepseek-v4-flash

Provider prices, cheapest first

DeepSeek · deepseek-v4-flash official price: input 0.3, output 1.2 USD per million tokens (DeepSeek, observed 2026-10-17). Cheapest inference provider: DeepInfra, input 0.09, output 0.18 USD per million tokens. 221 sites have comparable prices; 527 sites list this model.

Price ranges per million tokens

Official cash prices · Lowest 0.525 · median 0.525 · highest 0.525 · 1 quotes USD / per million tokens

Official cash prices · Lowest 3.5 · median 3.5 · highest 3.5 · 1 quotes CNY / per million tokens

Inference provider cash prices · Lowest 0.1125 · median 0.113125 · highest 0.11375 · 2 quotes USD / per million tokens

Official price

Provider	Input	Output	Cache read	Cache write	Unit	Observed	Conditions and source
DeepSeek	0.3	1.2	0.006	—	USD / 1M tokens	2026-10-17	Official source
DeepSeek	2	8	0.04	—	CNY / 1M tokens	2026-10-17	Official source

Official prices are the vendor's cash prices. Inference providers also list cash prices and can be compared directly. Relays list prices in site credit; what you actually pay depends on each site's top-up rate, so they cannot be compared with official prices directly.

Inference providers

Provider	Input	Output	Cache read	Cache write	Unit	Observed	Conditions and source	vs official
DeepInfra	0.09	0.18	0.018	—	USD / 1M tokens	2026-10-17	Official source	Input 70% lower · Output 85% lower
OpenRouter · Aggregator	0.091	0.182	0.0182	—	USD / 1M tokens	2026-10-17	Official source	Input 70% lower · Output 85% lower

Inference provider prices at least 10% above official (1)

Inference providers

Provider	Input	Output	Cache read	Cache write	Unit	Observed	Conditions and source	vs official
SiliconFlow	3.00	9.00	0.30	—	CNY / 1M tokens	2026-10-17	Official source	Input 50% higher · Output 13% higher

Unit: USD credits per 1M tokens. These are site-credit prices; actual cost depends on each site’s top-up rate.

Showing: Related variants。These are special editions of the model (thinking, max, search, batch, long context, free, etc.). They are priced differently from the base model, so they are listed separately and not ranked. Names that only add an effort level such as low, high or nothinking are compared with the base model. Back to the default ranking

Find a provider by name or address

Searching…

Browse all providers

Blended price = input × 75% + output × 25%, without cache.

Price range: lowest — · median — · highest — (0 quotes, USD credits per million total tokens)

sites have comparable prices221527 sites list this model; approximately 5 operators

Lowest input price—USD credits / 1M

Median of each provider’s lowest price—USD credits / 1M

Relays

Prices below are in site credit, not cash.

#	Provider	Plan / conditions	Input ↑USD credits / 1M	Output USD credits / 1M	Cache readUSD credits / 1M	BlendedUSD credits / 1M	Details
Related variants · not ranked · Recent price
—	xckjj.comAsk or discussCollected: 16 Oct 2026, 07:34 UTC Collected this cycle View 1 plans	Chinese models [machine translation]Group on site: 国模 Name on site: deepseek-v4-flash:free Related variant: deepseek-v4-flash:free (:free · free variant) Operator pending verification	0.0112	0.0224	0.000224	0.014	Full prices and sources
—	api.pezayo.comAsk or discussCollected: 16 Oct 2026, 07:41 UTC Collected this cycle View 2 plans	default Name on site: deepseek-v4-flash-thinking Default group, use all models via the openai chat completion format [machine translation]Original: 默认分组，通过 openai chat completion 格式使用全部模型 Related variant: deepseek-v4-flash-thinking (-thinking · thinking edition) Operator pending verification	0.143	0.286	0.00255999999999	0.17875	Full prices and sources
—	api.pezayo.comAsk or discussCollected: 16 Oct 2026, 07:41 UTC Collected this cycle View 2 plans	official Name on site: deepseek-v4-flash-thinking Official group, gpt, claude gemini official models [machine translation]Original: 官方分组，gpt, claude gemini 官方模型 Related variant: deepseek-v4-flash-thinking (-thinking · thinking edition) Operator pending verification	0.143	0.286	0.00255999999999	0.17875	Full prices and sources
—	yinli.oneAsk or discussCollected: 16 Oct 2026, 07:54 UTC Collected this cycle View 1 plans	deepseek Name on site: deepseek-v4-flash-max Deepseek channel [machine translation]Original: Deepseek渠道 Related variant: deepseek-v4-flash-max (-max · max edition) Operator pending verification	1	2	0.02	1.25	Full prices and sources
—	apinexus.dpdns.orgAsk or discussCollected: 16 Oct 2026, 07:43 UTC Collected this cycle View 2 plans	default Name on site: deepseek-v4-flash-max Default group [machine translation]Original: 默认分组 Related variant: deepseek-v4-flash-max (-max · max edition) Operator pending verification	60	180	—	90	Full prices and sources
—	apinexus.dpdns.orgAsk or discussCollected: 16 Oct 2026, 07:43 UTC Collected this cycle View 2 plans	vip Name on site: deepseek-v4-flash-max Full-price Pro group [machine translation]Original: 正价Pro分组 Related variant: deepseek-v4-flash-max (-max · max edition) Operator pending verification	90	270	—	135	Full prices and sources
Related variants · not ranked · Recent price · Includes prices of 0 (free status unconfirmed)
— · Price of 0 · free status unconfirmed	ai.bytonsec.comAsk or discussCollected: 16 Oct 2026, 07:35 UTC Collected this cycle View 2 plans	zen Name on site: zen/deepseek-v4-flash-free OPENCODE ZEN official 60% of list price [machine translation]Original: OPENCODE ZEN官方6折 Related variant: zen/deepseek-v4-flash-free (-free · free variant) Operator pending verification	0	0	—	0	Full prices and sources
— · Price of 0 · free status unconfirmed	ai.bytonsec.comAsk or discussCollected: 16 Oct 2026, 07:35 UTC Collected this cycle View 2 plans	zen Name on site: deepseek-v4-flash-free OPENCODE ZEN official 60% of list price [machine translation]Original: OPENCODE ZEN官方6折 Related variant: deepseek-v4-flash-free (-free · free variant) Operator pending verification	0	0	—	0	Full prices and sources

How this ranking works

All per-token prices: 97,455 prices from 871 sites (about 765 companies). 89,631 are ranked, from 861 sites (about 755 companies); 7,824 more are listed separately and not ranked. Prices collected up to 2026-10-17.

This model has 1052 recorded prices; 375 can be compared directly (354 ranked, 21 listed separately). Not comparable yet: tiered by rules (see Tiered prices) 291, price not parsed yet 244, billed per request or another unit 59, display unit not confirmed 36, ratio looks unset 26, scope of the official doc price not confirmed 8

8 recent · 0 stale (older than 10 days, or not updated by this refresh) · 0 unverified timestamps. No recent prices to summarize in this group. 2 zero-valued pairs are separately unverified, without proof of free eligibility. Timing does not verify payment, upstream identity or quality.

Prices are converted to site credits using each site’s published ratio. What you actually pay depends on its top-up rate. Input, output and cache prices in each row come from the same plan. Several sites run by the same company take one place in the ranking: its cheapest plan, with the other sites listed under it. We judge “same company” by identical prices, status pages or homepage code; this is not a legal finding. To see every site’s prices, choose “All plans” or filter by provider.

“Self-declared unofficial channel” means the plan name itself says “reverse-engineered” or similar. We have not verified their source or quality. Minimum and median only count regular plans.

Full methodology →

Switch model

Search modelsModel529 matches; choose and submit to view prices.

Sources not accepted this cycle and reasons

FAQ

Which API is cheapest for DeepSeek · deepseek-v4-flash?

No valid quotes within 10 days. Older quotes are listed as reference prices.

What is the official price for DeepSeek · deepseek-v4-flash?

Input 0.3, output 1.2 USD per million tokens, checked 2026-10-17. Lowest inference provider: DeepInfra, input 0.09, output 0.18 USD per million tokens. Input 70% lower than official.

How do I use DeepSeek · deepseek-v4-flash in Claude Code or Codex?

Install Grandet Skill, then say "Connect the cheapest DeepSeek · deepseek-v4-flash API to Claude Code for me". Install Skill →