Comparison · 9 min · 2 citations
DeepSeek vs Gemini API Pricing 2026: Per-Token Cost Compared
DeepSeek vs Gemini API pricing 2026: DeepSeek V4.1-Flash is $0.30/$1.20 per million tokens at peak, half off-peak; Gemini 3.5 Flash is $1.50/$9.00.
DeepSeek is far cheaper than Gemini's workhorse tier. DeepSeek V4.1-Flash is $0.30 input (cache miss) / $1.20 output per million tokens at peak and half that off-peak; Gemini 3.5 Flash is $1.50 / $9.00[1][2]. On output that is 7.5 times cheaper at peak and 15 times off-peak. Google's cheapest tier, Gemini 2.5 Flash-Lite at $0.10 / $0.40, now undercuts DeepSeek on both input and output.
DeepSeek wins on cost against Gemini Flash and Pro; Gemini buys data residency, documented long-context tiers, video and audio input, and Google Cloud integration. DeepSeek now charges by time of day, so the hour your jobs run sets half the bill.
DeepSeek and Gemini sit at opposite ends of the cost-versus-platform trade. DeepSeek V4.1-Flash prices text at $0.30 input / $1.20 output per million tokens at peak, undercutting Gemini 3.5 Flash ($1.50 / $9.00) several times over. This article lays out the verified rates, DeepSeek's peak-hour pricing and Gemini's context tiers, and runs both through the margin calculator so the cost gap shows up as gross margin on a real SaaS.
1. The per-token rate table
Standard rates per million tokens, verified against both official pricing pages: DeepSeek on October 2, 2026 (peak rates; off-peak is half), Gemini on May 25, 2026.
| Tier | DeepSeek model | Input / Output | Gemini model | Input / Output |
|---|---|---|---|---|
| Workhorse | DeepSeek V4.1-Flash[1] | $0.30 / $1.20 | Gemini 3.5 Flash[2] | $1.50 / $9.00 |
| Cheapest tier | DeepSeek V4.1-Flash, off-peak[1] | $0.15 / $0.60 | Gemini 2.5 Flash-Lite[2] | $0.10 / $0.40 |
| Cache hit (input) | V4.1-Flash cache hit[1] | $0.006 | Gemini 3.5 Flash cached[2] | $0.15 |
| Reasoning / pro | DeepSeek V4-Pro[1] | $1.32 / $3.96 | Gemini 3.1 Pro (≤200k)[2] | $2.00 / $12.00 |
| Pro, long context | One rate up to its 1M-token window | — | Gemini 3.1 Pro (>200k)[2] | $4.00 / $18.00 |
The headline gap is on output. DeepSeek V4.1-Flash output at $1.20 against Gemini 3.5 Flash at $9.00 is a 7.5x difference at peak, 15x off-peak. Against Google's cheapest tier the order flips: Gemini 2.5 Flash-Lite ($0.10 / $0.40) is below DeepSeek even off-peak ($0.15 / $0.60). At the pro tier DeepSeek V4-Pro ($1.32 / $3.96) costs about a third of Gemini 3.1 Pro on output.
2. DeepSeek's peak hours and cache rate
Two DeepSeek details change the real cost. First, time of day: peak rates apply from 01:00 to 04:00 and 06:00 to 10:00 UTC on weekdays, excluding Chinese public holidays, and every other hour bills at half[1]. A batch job scheduled outside those windows pays $0.15 / $0.60 on V4.1-Flash. Second, the cache-hit input rate of $0.006 per million tokens at peak ($0.003 off-peak) is a fiftieth of the cache-miss rate[1]; a workload that resends a large stable context pays close to nothing for it.
The model names changed with the repricing. DeepSeek's current models are DeepSeek-V4.1-Flash (API name deepseek-flash) and DeepSeek-V4-Pro-0813, both with a 1M-token context and 384K maximum output; the legacy deepseek-v4-flash name is still accepted and served by V4.1-Flash at the Flash price[1].
3. Gemini's tier ladder and context tiers
Gemini publishes a deeper tier ladder with context-dependent pricing[2]. Gemini 3.1 Pro charges $2.00 input / $12.00 output at or below 200k-token prompts, and $4.00 / $18.00 above 200k. Gemini 2.5 Pro has the same split shape ($1.25 / $10.00 at or below 200k, $2.50 / $15.00 above). The cheap tiers (2.5 Flash-Lite at $0.10 / $0.40, 3.1 Flash-Lite at $0.25 / $1.50) are where Gemini meets or beats DeepSeek on cost.
What the Gemini premium buys over DeepSeek: documented long-context pricing tiers, multimodal input priced per modality (text, image, video, audio) where DeepSeek V4.1-Flash takes text and images only, and Google Cloud data-residency and compliance options. For a text-only product on a mid-tier model, DeepSeek's cost advantage is large. For video or audio input, or EU data-residency requirements, Gemini's platform breadth can justify the higher rate.
4. Margin impact on a $19 SaaS
The margin calculator prices a $19/month SaaS where each user triggers 40 API calls a day at 800 input and 600 output tokens, first on DeepSeek V4.1-Flash at peak ($0.30 / $1.20), then on Gemini 3.5 Flash ($1.50 / $9.00). The engine renders gross margin from the verified rates:
Show the recompute-verified inputs and outputs
| subscription_price | 19 |
|---|---|
| avg_api_calls_per_day | 40 |
| avg_input_tokens | 800 |
| avg_output_tokens | 600 |
| input_cost_per_million | 0.3 |
| output_cost_per_million | 1.2 |
| hosting_cost_per_user | 0.5 |
| other_per_user_costs | 0.25 |
| api cost per user | 1.15 |
|---|---|
| total cost per user | 1.9 |
| gross margin per user | 17.1 |
| gross margin percent | 90 |
| api share percent | 60.5 |
| dominant cost driver | AI API |
| scale tiers › row 1 › users | 100 |
| scale tiers › row 1 › total revenue | 1900 |
| scale tiers › row 1 › total cost | 190 |
| scale tiers › row 1 › total profit | 1710 |
| scale tiers › row 1 › margin percent | 90 |
| scale tiers › row 2 › users | 1000 |
| scale tiers › row 2 › total revenue | 19000 |
| scale tiers › row 2 › total cost | 1900 |
| scale tiers › row 2 › total profit | 17100 |
| scale tiers › row 2 › margin percent | 90 |
| scale tiers › row 3 › users | 10000 |
| scale tiers › row 3 › total revenue | 190000 |
| scale tiers › row 3 › total cost | 19000 |
| scale tiers › row 3 › total profit | 171000 |
| scale tiers › row 3 › margin percent | 90 |
| insight | AI API is 60.5% of your per-user cost ($1.15 a month). At 10K users, that is $11,500 a month on API alone. Caching responses or using a tiered model approach could significantly improve margins. |
Computed live at build time.
| subscription_price | 19 |
|---|---|
| avg_api_calls_per_day | 40 |
| avg_input_tokens | 800 |
| avg_output_tokens | 600 |
| input_cost_per_million | 1.5 |
| output_cost_per_million | 9 |
| hosting_cost_per_user | 0.5 |
| other_per_user_costs | 0.25 |
| api cost per user | 7.92 |
|---|---|
| total cost per user | 8.67 |
| gross margin per user | 10.33 |
| gross margin percent | 54.4 |
| api share percent | 91.3 |
| dominant cost driver | AI API |
| scale tiers › row 1 › users | 100 |
| scale tiers › row 1 › total revenue | 1900 |
| scale tiers › row 1 › total cost | 867 |
| scale tiers › row 1 › total profit | 1033 |
| scale tiers › row 1 › margin percent | 54.4 |
| scale tiers › row 2 › users | 1000 |
| scale tiers › row 2 › total revenue | 19000 |
| scale tiers › row 2 › total cost | 8670 |
| scale tiers › row 2 › total profit | 10330 |
| scale tiers › row 2 › margin percent | 54.4 |
| scale tiers › row 3 › users | 10000 |
| scale tiers › row 3 › total revenue | 190000 |
| scale tiers › row 3 › total cost | 86700 |
| scale tiers › row 3 › total profit | 103300 |
| scale tiers › row 3 › margin percent | 54.4 |
| insight | AI API is 91.3% of your per-user cost ($7.92 a month). At 10K users, that is $79,200 a month on API alone. Caching responses or using a tiered model approach could significantly improve margins. |
Computed live at build time.
The gap between the two runs is large: on DeepSeek the API costs about $1.15 a user a month at peak and the product keeps a 90% gross margin, while Gemini 3.5 Flash costs about $7.92 a user and the margin falls to roughly 54%. At 40 calls a day with 600 output tokens each, the output rate ($1.20 against $9.00) dominates everything. For a high-call-volume product on a low subscription price, the model choice is the difference between a healthy margin and a thin one.
The honest caveat: the calculator prices cost, not capability. If Gemini 3.5 Flash produces materially better output on your task, the higher cost may pay for itself in retention. And if cost is all that matters, Gemini 2.5 Flash-Lite is cheaper still than either. To compare against Western flagships, see Anthropic vs OpenAI pricing and the full cheapest LLM API ranking.
5. Decision guidance
- Text and images, high call volume, mid-tier quality: DeepSeek V4.1-Flash, off-peak where the work can wait. Output at $1.20 (peak) against Gemini 3.5 Flash's $9.00 decides it.
- Lowest possible cost: Gemini 2.5 Flash-Lite, below DeepSeek even at DeepSeek's off-peak rate.
- Video or audio input: Gemini, which prices each modality; DeepSeek V4.1-Flash takes text and images only.
- EU data residency or Google Cloud integration: Gemini, where data-residency options are documented.
- Reasoning workloads on a budget: DeepSeek V4-Pro at $1.32 / $3.96 peak, about a third of Gemini 3.1 Pro's output rate.
Re-verify both pricing pages before committing; DeepSeek has repriced twice in 2026 and now prices by time of day. For the full provider field including open-weight hosts, see the cheapest LLM API ranking.
Per-token figures verified against official pricing pages: DeepSeek on 2026-10-02, Gemini on 2026-05-25.
Frequently asked questions
Is DeepSeek or Gemini cheaper in 2026?
It depends on the tier. Against Gemini 3.5 Flash ($1.50 input / $9.00 output per million tokens), DeepSeek V4.1-Flash is far cheaper: $0.30 input and $1.20 output at peak, half that off-peak, so 7.5 to 15 times cheaper on output. Against Google's cheapest tier, Gemini 2.5 Flash-Lite at $0.10 / $0.40, DeepSeek is now the more expensive of the two even off-peak ($0.15 / $0.60). DeepSeek rates verified on its pricing docs on October 2, 2026.
When are DeepSeek's off-peak hours?
DeepSeek bills peak rates from 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday, excluding Chinese public holidays. Every other hour, weekends included, is off-peak at half the peak rate. A batch job that can wait for off-peak hours pays $0.15 input and $0.60 output per million tokens on V4.1-Flash.
Why would I pay more for Gemini over DeepSeek?
Gemini buys Google's data-residency options, documented long-context tiers above 200k tokens, multimodal input across text, image, video and audio, and integration with Google Cloud. DeepSeek V4.1-Flash accepts images but not video or audio. The choice is cost against platform and modality breadth, not a pure price decision.
References
Sources
Primary sources only. No vendor-marketing blogs or aggregated secondary claims.
- 1 DeepSeek — API pricing (V4.1-Flash and V4-Pro cache-hit, cache-miss and output rates; peak hours) — accessed 2026-10-02
- 2 Google — Gemini API pricing (3.5 Flash, 3.1 Pro, 3.1 Flash-Lite, 2.5 series per-MTok rates) — accessed 2026-05-25
Tools referenced in this article