deepseek/deepseek-chat-v3-0324:free
open_router/deepseek/deepseek-chat-v3-0324:free
Context
—
Max output
16K
Input / 1M
Free
Output / 1M
Free
Modality
Chat
Cutoff
Aug 2024
About this model
DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.
Best suited for
- General-purpose chat, content generation, basic coding help, and developer experimentation in community or free-tier environments.
Capabilities
System prompt
Honours a dedicated system role, separate from the user turn.
Supported parameters
creativity_levelControls the creativity of responses. Higher values (e.g., 0.7) increase creativity; lower values (e.g., 0.2) make responses more predictable.
max_tokensMax Tokens LimitSpecifies the maximum number of text units (tokens) allowed in a response, limiting its length.
probability_cutoffProbability Cutoff (Top P)Focuses on the most likely words based on a percentage of probability.
log_probabilityIf true, returns the log probabilities of each output token returned in the content of message.
repetition_penaltyThe `frequency_penalty` controls how often the model repeats itself, with higher positive values reducing repetition and negative values encouraging it.
novelty_penaltyDiscourages responses that are too similar to previous ones.
stopThis parameter tells the model to stop generating text when it reaches any of the specified sequences (like a word or punctuation)
response_typeDefines the format or type of the generated response.
parallel_tool_callsEnables parallel execution of tools, allowing multiple tools to run simultaneously.
streamSends the response in real-time as it's being generated.