DeepSeek
V4.1 Flash now lowers the direct API price, while V4 Pro remains available after DeepSeek reversed its planned September 14 reroute. Microsoft Foundry still exposes the older V4 Flash meters with resale markups from ~10% to a reported 4.5x.
DeepSeek V4.1 Flash and V4 Pro ship real 1M-token windows (max output up to 384K). The direct `deepseek-flash` endpoint now serves V4.1 Flash with lower peak/off-peak rates, while legacy V4 Flash names route to the same model. Peak hours are weekdays excluding Chinese public holidays; the published 2026 holiday dates are modeled explicitly, while future annual calendars will need updating when China publishes them.
Pricing remains a resale case study: the direct API is cheap, Microsoft Foundry resells it at a markup, and some Foundry tiers bill a cache meter the public page hides. DeepSeek's current API Quick Start and pricing pages now say V4 Pro service continues after 2026-09-14 with billing unchanged, although the September 10 launch article still carries the earlier reroute notice. The catalog therefore retains V4 Pro's separate direct lane and schedule; the Foundry V4 Flash snapshot remains separate, while the direct Flash row carries the September 10 transition and both published schedules.
Pricing
per 1M tokens · USD / CHF| Model | Tier / Host | Context | Input | Cached | Output | Confidence |
|---|---|---|---|---|---|---|
DeepSeek-V4 Pro 1M context. Cache meter is published in Azure's retail catalog. | Foundry ·Global | 1M 384K max out | $1.74CHF 1.40 | $0.145CHF 0.117 | $3.48CHF 2.80 | official Azure retail catalog: $1.74/M input, $0.145/M cached input, and $3.48/M output. |
DeepSeek-V4 Flash 1M context. Cache meter is published in Azure's retail catalog. | Foundry ·Global | 1M 384K max out | $0.19CHF 0.153 | $0.028CHF 0.023 | $0.51CHF 0.411 | official Azure retail catalog: $0.19/M input, $0.028/M cached input, and $0.51/M output. |
DeepSeek-V4 Pro First-party published cache-hit, cache-miss, and output rates after the 75% direct price cut. Peak/off-peak billing began 2026-08-16 16:00 UTC (peak 01:00-04:00 & 06:00-10:00 UTC, Monday to Friday, excluding Chinese public holidays): off-peak $0.022/$0.66/$1.98, peak $0.044/$1.32/$3.96 — see the changelog for detail. DeepSeek's current API docs say this model remains available after 2026-09-14 with billing unchanged. | Direct API DeepSeek direct API | 1M | $0.66CHF 0.531 | $0.022CHF 0.018 | $1.98CHF 1.59 | official DeepSeek's own direct (non-cloud-resold) API pricing, including the published cached-input rate. The API Quick Start and Models & Pricing pages, re-read 2026-09-12, state that V4 Pro service continues after September 14, 2026 with billing unchanged; the September 10 launch article still carries the earlier automatic-reroute notice, so the catalog keeps this V4 Pro lane separate and documents the conflict. |
Off-peakChanges in 12h 43m Peak$1.32 / $0.044 / $3.96Off-peak$0.66 / $0.022 / $1.98 | ||||||
DeepSeek-V4.1 Flash The direct `deepseek-flash` alias now serves the `DeepSeek-V4.1-Flash-1017` snapshot, upgraded October 17; the API call method and current peak/off-peak price schedule are unchanged. Before 2026-09-10 04:00 UTC, this row's base and legacy variants preserve the V4 Flash schedule; from the transition onward, off-peak is $0.15/$0.003/$0.60 and weekday peak (except Chinese public holidays) is $0.30/$0.006/$1.20 per M (cache hit / cache miss / output order is shown in the source table; the catalog fields are input / cached input / output). | Direct API DeepSeek direct API | 1M 384K max out | $0.15CHF 0.121 | $0.003CHF 0.002 | $0.60CHF 0.483 | official DeepSeek's official API changelog, V4.1 Flash-1017 announcement, and Models & Pricing page, captured 2026-10-17, identify `deepseek-flash` as the alias for the new DeepSeek-V4.1-Flash-1017 snapshot and state that the API call method and pricing are unchanged. The live table retains the 1M context, 384K maximum output, and the peak/off-peak rates represented by the variants below. The base fields retain the pre-transition V4 Flash off-peak rate for deterministic historical resolution. |
Off-peakChanges in 12h 43m Peak$0.30 / $0.006 / $1.20Off-peak$0.15 / $0.003 / $0.60Peak (legacy V4 Flash)$0.44 / $0.014 / $1.32Off-peak (legacy V4 Flash)$0.22 / $0.007 / $0.66 | ||||||
DeepSeek-V4.1 Flash Microsoft Foundry Direct from Azure public preview, billed through the Azure subscription. The flat Global Standard price is separate from the DeepSeek API's time-based variants. Azure's retail feed also emits Data Zone meters, but Microsoft's launch table documents Global Standard only, so no Data Zone lane is inferred. | Foundry ·Global Direct from Azure | 1M | $0.30CHF 0.241 | $0.006CHF 0.005 | $1.20CHF 0.966 | official Microsoft Foundry Blog (techcommunity.microsoft.com/blog/azure-ai-foundry-blog/deepseek-v4-1-flash-is-coming-to-microsoft-foundry/4556431) lists Direct from Azure Global Standard at $0.30/M input, $0.006/M cache, and $1.20/M output. Azure Retail Prices API, serviceName 'Foundry Models', productName 'Azure Deepseek Models', captured 2026-10-09, now shows matching Global meters effective 2026-10-01. It also emits Data Zone meters, but Microsoft's launch table documents Global Standard only; those Data Zone rows are not modeled until Microsoft confirms deployment support. The Azure price has no peak/off-peak schedule, unlike DeepSeek's Direct API. |
DeepSeek-V4.1 Flash New Fireworks-hosted Microsoft Foundry Global lane. Its rates are distinct from both DeepSeek's direct API and Fireworks' own direct endpoint. | Foundry ·Global Fireworks-hosted | 1M | $0.375CHF 0.302 | $0.008CHF 0.006 | $1.50CHF 1.21 | official Azure Retail Prices API query for product 'Azure Fireworks Models' filtered to V4.1-Flash meters, captured 2026-09-24: effective 2026-09-01 meters 'FW DS-V4.1-Flash Gl Inp Tokens' $0.000375/1K ($0.375/M), 'FW DS-V4.1-Flash Gl Cd Inp Tokens' $0.000008/1K ($0.008/M), and 'FW DS-V4.1-Flash Gl Opt Tokens' $0.0015/1K ($1.50/M), uniform across 20 commercial Global regions. Microsoft's Sep 23 Foundry announcement rounds the first two values to $0.37 and $0.007 per M; this catalog keeps the more precise billable Retail rates. Fireworks' official model page identifies V4.1-Flash as a ready serverless model; this is the separately metered Foundry resale lane. |
DeepSeek-V4 Pro Third-party inference provider's direct rate. | Direct API Fireworks direct API | — | $1.74CHF 1.40 | — | $3.48CHF 2.80 | official Fireworks direct API pricing. |
DeepSeek-V4 Pro First-party Foundry Data Zone deployment — slightly cheaper than the Fireworks-hosted Data Zone lane below ($1.925/$0.165/$3.828). | Foundry ·Data Zone | 1M | $1.91CHF 1.54 | $0.16CHF 0.129 | $3.83CHF 3.08 | official Azure Retail Prices API, product "Azure Deepseek Models": 'V4 Pro Inp DZ Tokens' $0.00191/1K, 'V4 Pro cached DZ Tokens' $0.00016/1K (effective 2026-07-01), 'V4 Pro Outp DZ Tokens' $0.00383/1K, consistent across 22 commercial regions. Captured 2026-07-27. |
DeepSeek-V4 Pro ~10% above the Fireworks direct rate: the Data Zone premium. This is the Fireworks-hosted lane; a cheaper first-party Foundry Data Zone deployment also exists (see above). | Foundry ·Data Zone Fireworks-hosted | — | $1.93CHF 1.55 | $0.165CHF 0.133 | $3.83CHF 3.08 | official Fireworks official live pricing page. |
DeepSeek-V4 Flash First-party Foundry Data Zone deployment: the Global rate x1.10, rounded up to the meter's precision on every dimension. The Fireworks-hosted Data Zone lane below undercuts both this and Global. | Foundry ·Data Zone | 1M 384K max out | $0.21CHF 0.169 | $0.031CHF 0.025 | $0.56CHF 0.451 | official Azure Retail Prices API, product "Azure Deepseek Models": 'V4 Flash Inp DZ Tokens' $0.00021/1K, 'V4 Flash cached DZ Tokens' $0.000031/1K, 'V4 Flash Outp DZ Tokens' $0.00056/1K, uniform across 22 commercial regions. Captured 2026-08-22. Each figure is the Global rate x1.10 rounded up ($0.19 to $0.209 to $0.21; $0.028 to $0.0308 to $0.031; $0.51 to $0.561 to $0.56). The two US-Gov regions price higher, as they do across this whole product; the commercial majority is used here. Microsoft separately publishes a 'V4 Flash 0731' meter set for the newer snapshot at $0.44/$0.014/$1.32 Global — this row tracks the plain 'V4 Flash' meters. |
DeepSeek-V4 Flash The cheapest DeepSeek lane on Foundry, and cheaper than the Global tier — an inversion of the usual pattern, because this is a different host undercutting the first-party listing rather than a tier discount. | Foundry ·Data Zone Fireworks-hosted | — | $0.15CHF 0.121 | $0.03CHF 0.024 | $0.31CHF 0.25 | official Azure Retail Prices API, product "Azure Fireworks Models": 'FW Deepseek-v4-Flash In DZ Tokens' $0.00015/1K, 'FW Deepseek-v4-Flash Cd In DZ Tokens' $0.00003/1K, 'FW Deepseek-v4-Flash Opt DZ Tokens' $0.00031/1K, effective 2026-08-01, uniform across all 20 commercial Data Zone regions with no US-Gov or rounding outlier. Captured 2026-08-22. Fireworks prices its own hosting, so this is not a multiple of DeepSeek's or Microsoft's rate, and the meter names no model snapshot. |
Quirks & gotchas
V4 Pro remains available after the planned switch
DeepSeek's current API Quick Start and Models & Pricing pages say V4 Pro service continues after September 14, 2026, with billing unchanged. The September 10 V4.1 Flash launch article still says V4 Pro requests would route to Flash at Flash rates from that date, so the official pages are inconsistent; the catalog keeps the published V4 Pro lane and its own peak/off-peak schedule until DeepSeek clarifies the conflict.
Resale markup: ~10% to a reported 4.5x
Direct V4 Pro is $0.435 / CHF 0.35 input. A "Global" Microsoft Foundry V4 Pro has been reported at about 4.5x that, the widest markup here. The Fireworks Data Zone listing ($1.925 / CHF 1.55) is a milder ~10% over Fireworks direct ($1.74 / CHF 1.40).
Azure cache meters are publicly priced
The Azure pricing summary does not show a cached-input column, but Azure's retail catalog lists one for both Global models: $0.145 / CHF 0.12 per M for V4 Pro (~91.7% off) and $0.028 / CHF 0.02 for V4 Flash (~85% off). Billing exports reconcile to those published meters.
V4 Flash: Data Zone can be cheaper than Global
Data Zone is normally the premium tier — it pins routing to a geography and charges about 10% for it. V4 Flash breaks that. Microsoft's own Data Zone deployment does follow the rule ($0.21 / CHF 0.17 input against $0.19 / CHF 0.15 Global), but the Fireworks-hosted Data Zone listing bills $0.15 / CHF 0.12 input and $0.31 / CHF 0.25 output — below the Global tier on every dimension, and the cheapest DeepSeek lane on Foundry.
The reason is that these are different sellers, not different tiers of one seller. Comparing tier labels across hosts tells you nothing about price; compare the meters. If you were going to accept Data Zone routing anyway, the Fireworks lane is strictly cheaper than staying on Global.
1M context is a real autonomy lever
V4 Pro and V4.1 Flash carry 1M-token windows (max output up to 384K). Bigger windows mean less forced compaction, so less babysitting.
No first-party embedding model
The chat models are generation-only. A separate deepseek-embedding-v2 (768-dim) exists, but for code RAG the Embeddings page recommends Cohere embed-v4.