Scheduled for retirement on Mar 30, 2027. Still available at the prices below until then.
Context
164K
Max output
164K
Serving providers
11
Cheapest input
$0.25 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial AnalysisSpeed & latency
Output speed
94 t/s
tokens / second
Time to first token
1.44s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CritPt | 2% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 127.6 | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| GPQA Diamond | 78% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| Humanity's Last Exam | 14% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| IFBench | 41% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| LiveCodeBench | 78% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| Long-Context Reasoning | 57% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| MMLU-ProFrozen Aug 18, 2026 | 85% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| Omniscience | -29.6 | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| Omniscience Accuracy | 0.3 | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| Omniscience Non Hallucination | 0.2 | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| SciCodeArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 39% | — | — | |||||||||||||
| ||||||||||||||||
| TerminalBench Hard | 25% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
| τ²-bench | 37% | reasoning: true | — | |||||||||||||
| ||||||||||||||||
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.25 to $3.00 per 1M tokens across 11 providers, so the dearest route costs 1100% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfraCheapest | $0.25 | $0.95 | 1 | |||||||||||||
Standarddeepinfra/deepseek-ai/DeepSeek-V3.1
| ||||||||||||||||
| VercelVercel reseller quote | $0.25 | $0.95 | 1 | |||||||||||||
Standarddeepseek/deepseek-v3.1
| ||||||||||||||||
| Novita | $0.27 | $1.00 | 1 | |||||||||||||
Standardnovita/deepseek/deepseek-v3.1
| ||||||||||||||||
| Baseten | $0.50 | $1.50 | 1 | |||||||||||||
Standardbaseten/deepseek-ai/DeepSeek-V3.1
| ||||||||||||||||
| Wandb | $0.55 | $1.65 | 1 | |||||||||||||
Standardwandb/deepseek-ai/DeepSeek-V3.1
| ||||||||||||||||
| Fireworks AI | $0.56 | $1.68 | 1 | |||||||||||||
Standardfireworks_ai/accounts/fireworks/models/deepseek-v3p1
| ||||||||||||||||
| Amazon Bedrock | $0.58 | $1.68 | 1 | |||||||||||||
Standardbedrock_mantle/deepseek.v3.1
| ||||||||||||||||
| Together AI | $0.60 | $1.70 | 1 | |||||||||||||
Standardtogether_ai/deepseek-ai/DeepSeek-V3.1
| ||||||||||||||||
| SambaNova is dearest at $3.00 / $4.50 | ||||||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Input down 22% since first tracked
Cost calculator
estimate your monthly spend on this model$126
estimated / month
Model IDs
copy the exact identifier for your platformdeepinfra/deepseek-ai/DeepSeek-V3.1deepseek/deepseek-v3.1novita/deepseek/deepseek-v3.1baseten/deepseek-ai/DeepSeek-V3.1wandb/deepseek-ai/DeepSeek-V3.1fireworks_ai/accounts/fireworks/models/deepseek-v3p1bedrock_mantle/deepseek.v3.1together_ai/deepseek-ai/DeepSeek-V3.1replicate/deepseek-ai/deepseek-v3.1azure_ai/deepseek-v3.1sambanova/DeepSeek-V3.1
Frequently asked questions
DeepSeek V3.1 pricing, context and availabilityIs DeepSeek V3.1 still available?
Yes. DeepSeek V3.1 is scheduled for retirement on 2027-03-30.
How much does DeepSeek V3.1 cost?
DeepSeek V3.1 costs $0.25 per 1M input tokens from DeepInfra, and $0.95 per 1M output tokens from DeepInfra. Across 11 serving providers, input prices range from $0.25 to $3.00 per 1M tokens.
What is the context window of DeepSeek V3.1?
DeepSeek V3.1 accepts up to 164K tokens of context and can return up to 164K output tokens.
Which providers serve DeepSeek V3.1?
DeepSeek V3.1 is available from 11 serving providers, each with its own pricing and model ID. DeepInfra is currently the cheapest.
What can DeepSeek V3.1 do?
DeepSeek V3.1 supports tool use, reasoning and structured output.
Other DeepSeek models
compare pricing across the DeepSeek lineupDeepSeek V4 Flash 0423$0.09/1M · 18 providersDeepSeek V4 Flash 0731$0.016/1M · 16 providersDeepSeek V4 Pro 0423$0.435/1M · 16 providersR1$0.40/1M · 14 providersDeepSeek V4.1 Flash$0.01/1M · 12 providersDeepSeek V3 0324$0.20/1M · 12 providersDeepSeek V4 Pro 0813$0.66/1M · 12 providersDeepSeek R1 Distill Llama 70B$0.20/1M · 11 providers
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI