Thank you for choosing our service. Add credits today to receive an additional 20% bonus. Add Credits
DeepSeek: DeepSeek V3.1
deepseek/deepseek-chat-v3.1
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference.
The model improves tool use, code generation, and reasoning efficiency, achieving performance comparable to DeepSeek-R1 on difficult benchmarks while responding more quickly. It supports structured tool calling, code agents, and search agents, making it suitable for research, coding, and agentic workflows.
It succeeds the DeepSeek V3-0324 model and performs well on a variety of tasks.
Recent activity on DeepSeek V3.1
Tokens processed per day
Thoughput
(tokens/s)
| Providers | Min (tokens/s) | Max (tokens/s) | Avg (tokens/s) |
|---|---|---|---|
| Theta | 14.05 | 81.82 | 52.61 |
| Volcengine | 14.67 | 41.3 | 23.38 |
First Token Latency
(ms)
| Providers | Min (ms) | Max (ms) | Avg (ms) |
|---|---|---|---|
| Theta | 570 | 9357 | 1606.59 |
| Volcengine | 882 | 2354 | 1257.34 |
Providers for DeepSeek V3.1
ZenMux Provider to the best providers that are able to handle your prompt size and parameters, with fallbacks to maximize uptime.
Latency
0.62
s
Throughput
16.41
tps
Uptime
100.00
%
Recent uptime
Oct 10,2025 - 3 PM100.00%
Price
Input
$ 0.28
/ M tokens
Output
$ 1.11
/ M tokens
Cache read
-
Cache write 5m
-
Cache write 1h
-
Cache write
-
Web search
-
Model limitation
Context
128.00K
Max output
65.54K
Supported Parameters
max_completion_tokens
temperature
top_p
frequency_penalty
presence_penalty
seed
-
logit_bias
-
logprobs
-
top_logprobs
-
response_format
stop
tools
tool_choice
parallel_tool_calls
-
Model Protocol Compatibility
openai
anthropic
-
Data policy
Prompt training
false
Prompt Logging
Zero retention
Moderation
Responsibility of developer
Latency
-
Throughput
-
Uptime
100.00
%
Recent uptime
Oct 10,2025 - 3 PM100.00%
Price
Input
$ 0.56
/ M tokens
Output
$ 1.68
/ M tokens
Cache read
$ 0.11
/ M tokens
Cache write 5m
-
Cache write 1h
$ 0.0024
/ M tokens
Cache write
-
Web search
-
Model limitation
Context
128.00K
Max output
65.54K
Supported Parameters
max_completion_tokens
temperature
top_p
frequency_penalty
presence_penalty
seed
-
logit_bias
-
logprobs
-
top_logprobs
-
response_format
stop
tools
tool_choice
parallel_tool_calls
-
Model Protocol Compatibility
openai
anthropic
-
Data policy
Prompt training
false
Prompt Logging
Zero retention
Moderation
Responsibility of developer
Sample code and API for DeepSeek V3.1
ZenMux normalizes requests and responses across providers for you.
OpenAI-PythonPythonTypeScriptOpenAI-TypeScriptcURL
pythonfrom openai import OpenAI client = OpenAI( base_url="https://zenmux.ai/api/v1", api_key="<ZenMux_API_KEY>", ) completion = client.chat.completions.create( model="deepseek/deepseek-chat-v3.1", messages=[ { "role": "user", "content": [ { "type": "text", "text": "What is in this image?" } ] } ] ) print(completion.choices[0].message.content)