impulse

DeepSeek V3.1

Open weights
API id DeepSeek-V3.1
Input / 1M
$0.25
at DeepInfra
Output / 1M
$0.95
at DeepInfra
Context window
164K
tokens
Max output
164K
tokens per reply
Hosts
12
12 listings
Released
-
date not published

Pricing

List prices at DeepInfra, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.25$0.95
Cached input
48% off input
$0.13

What it costs a month

At DeepInfra list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$135
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$125
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$256
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$828
Use your own numbers

Where to run it

12 hosts serve DeepSeek V3.1. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
DeepInfraInference
DeepSeek-V3.1
$0.25 / $0.95
GMI CloudInference
DeepSeek-V3.1
$0.27 / $1.00
+6% vs headline
Novita AIInference
deepseek-v3.1
$0.27 / $1.00
+6% vs headline
BasetenInference
DeepSeek-V3.1
$0.50 / $1.50
+76% vs headline
W&B InferenceInference
DeepSeek-V3.1
$0.50 / $1.50
+76% vs headline
Fireworks AIInference
deepseek-v3p1
$0.56 / $1.68
+98% vs headline
deepseek-v3.1
$0.571 / $1.71
+102% vs headline
deepseek-v3.1-maas
$0.60 / $1.70
+106% vs headline
Together AIInference
DeepSeek-V3.1
$0.60 / $1.70
+106% vs headline
ReplicateInference
deepseek-v3.1
$0.672 / $2.02
+137% vs headline
deepseek-v3.1
$1.23 / $4.94
+408% vs headline
SambaNovaInference
DeepSeek-V3.1
$3.00 / $4.50
+694% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: DeepSeek License (MIT code)
Runtimes: SGLang, TGI, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than DeepInfra.

Smaller DeepSeek models

No cheaper DeepSeek model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    53% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    53% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Lacks reasoning
    38% less

Open-weight picks

  • MiniMax M3
    MiniMax · on EmpirioLabs AI
    $0.225 / $0.90
    Has every key capability
    7% less
  • Qwen3.5 397B A17B
    Alibaba Qwen · on AIHubMix
    $0.164 / $0.986
    Has every key capability
    13% less
  • Muse Glimmer 30B
    Meta · on EmpirioLabs AI
    $0.20 / $0.80
    Has every key capability
    18% less
  • Mistral Small 2603
    Mistral AI · on Mistral AI
    $0.15 / $0.60
    Has every key capability
    38% less

Model line

Newer and older versions, at each lab's own price.

  1. UndatedDeepSeek V3.2.251201LatestNot listed
  2. Dec 2025DeepSeek V3.2$0.28 / $0.40
  3. UndatedDeepSeek V3.1This page$0.25 / $0.95
  4. UndatedDeepSeek V3$0.89 / $0.89
  5. UndatedDeepSeek V2.5$1.20 / $1.20

Call it

Using DeepInfra's API.

Python
from litellm import completion

response = completion(
    model="deepinfra/DeepSeek-V3.1",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does DeepSeek V3.1 cost?

On DeepInfra, DeepSeek V3.1 costs $0.25 per million input tokens and $0.95 per million output tokens. Cached input costs $0.13 per million tokens.

Which APIs offer DeepSeek V3.1?

DeepSeek V3.1 runs on 12 APIs: DeepInfra, GMI Cloud, Novita AI, Baseten, W&B Inference, Fireworks AI, StreamLake, Vertex AI, Together AI and 3 more.

What is DeepSeek V3.1's context window?

164K tokens, with up to 164K tokens in one reply.

Is DeepSeek V3.1 being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of DeepSeek V3.1?

Yes. DeepSeek V3.2.251201 is the newest model in this line.

What are cheaper alternatives to DeepSeek V3.1?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (53% less), GPT-6 Luna on OpenAI (53% less) and MiniMax M3 on EmpirioLabs AI (7% less). Savings are on a mix of 3 input tokens to 1 output token.

What is DeepSeek V3.1's API model id?

Use DeepSeek-V3.1 with DeepInfra. Other hosts use their own ids: deepseek-v3.1 on Novita AI, deepseek-v3p1 on Fireworks AI, deepseek-v3.1 on StreamLake, deepseek-v3.1-maas on Vertex AI.

Sources

Raw catalog entry for DeepSeek-V3.1 on DeepInfra
{
  "max_input_tokens": 163840,
  "max_output_tokens": 163840,
  "input_cost_per_token": 2.5e-7,
  "output_cost_per_token": 9.5e-7,
  "cache_read_input_token_cost": 1.3e-7,
  "litellm_provider": "deepinfra",
  "mode": "chat",
  "supports_tool_choice": true,
  "supports_reasoning": true,
  "supports_function_calling": true,
  "source": "https://deepinfra.com/pricing"
}