Context
1M
Max output
384K
Serving providers
4
Cheapest input
$0.43 /1M
Prices are on-demand list rates. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial Analysis and Hugging Face leaderboardsSpeed & latency
Output speed
75 t/s
tokens / second
Time to first token
7.80s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AA-Briefcase | 22% | — | — | |||||||||||||||||
| APEX Agents | 24% | default | — | |||||||||||||||||
| CritPt | 13% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| EvalComputeProxy | 1422.7 | high | — | |||||||||||||||||
| ||||||||||||||||||||
| GDPval | 1306.1 | default | — | |||||||||||||||||
| ||||||||||||||||||||
| GPQA Diamond | 91% | high | — | |||||||||||||||||
| ||||||||||||||||||||
| Humanity's Last Exam | 36% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| IFBench | 76% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| IT-Bench SRE | 38% | default | — | |||||||||||||||||
| Long-Context Reasoning | 66% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| Omniscience | -9.7 | high | — | |||||||||||||||||
| ||||||||||||||||||||
| SciCode | 50% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| SWE-bench Pro | 55% | default | — | |||||||||||||||||
| SWE-bench Verified | 81% | default | — | |||||||||||||||||
| TerminalBench Hard | 46% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| TerminalBench v2.1 | 64% | default | — | |||||||||||||||||
| τ-bench Banking | 26% | default | — | |||||||||||||||||
| τ²-bench | 96% | default | — | |||||||||||||||||
| ||||||||||||||||||||
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.43 to $1.74 per 1M tokens across 4 providers, so the dearest route costs 300% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | ||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TencentCheapest | $0.43 | $0.87 | 1 | ||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||
| DeepSeekCheapest | $0.43 | $0.87 | 1 | ||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||
| Azure | $1.74 | $3.48 | 1 | ||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||
| Fireworks AI | $1.74 | $3.48 | 1 | ||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||
Price history
input + output $/1M since we started tracking Input Output
Input down 75% since first tracked
Cost calculator
estimate your monthly spend on this model$157
estimated / month
Model IDs
copy the exact identifier for your platformtencent/deepseek-v4-prodeepseek-v4-proazure_ai/deepseek-v4-profireworks_ai/accounts/fireworks/models/deepseek-v4-pro
Frequently asked questions
DeepSeek V4 Pro pricing, context and availabilityHow much does DeepSeek V4 Pro cost?
DeepSeek V4 Pro costs $0.43 per 1M input tokens and $0.87 per 1M output tokens at its cheapest provider via Tencent. Across 4 serving providers, input prices range from $0.43 to $1.74 per 1M tokens.
What is the context window of DeepSeek V4 Pro?
DeepSeek V4 Pro accepts up to 1M tokens of context and can return up to 384K output tokens.
Which providers serve DeepSeek V4 Pro?
DeepSeek V4 Pro is available from 4 serving providers, each with its own pricing and model ID. Tencent is currently the cheapest.
What can DeepSeek V4 Pro do?
DeepSeek V4 Pro supports tool use, reasoning, prompt caching and structured output.
Other DeepSeek models
compare pricing across the DeepSeek lineupPopular comparisons
head-to-head pages featuring DeepSeek V4 Pro
DeepSeek V4 Pro vs Claude Fable 5side-by-side pricing & benchmarks DeepSeek V4 Pro vs Claude Opus 4.8side-by-side pricing & benchmarks DeepSeek V4 Pro vs Claude Sonnet 5side-by-side pricing & benchmarks
DeepSeek V4 Pro vs Gemini 3.1 Pro Previewside-by-side pricing & benchmarks DeepSeek V4 Pro vs Gemini 3.5 Flashside-by-side pricing & benchmarks Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI