← Back to all generators

qwen/qwen3-7-plus

Qwen3.7-Plus is Alibaba's cost-effective multimodal model with vision-language understanding, a 1 million token context window, and strong agentic coding and tool use.

Capabilities

Reference ImagesSystem PromptMax TokensTop-P

Cost

Community model (estimated from hardware time)

Input Parameters

frequency_penaltynumber

Frequency penalty. Positive values penalize tokens proportionally to how often they have appeared, reducing repetition.

Default: 0min: -2, max: 2
imagearray

Optional list of input images for visual understanding (Qwen3.7-Plus is multimodal). Each image is sent alongside the text prompt.

Default:
max_tokensinteger

The maximum number of tokens the model should generate as output.

Default: 2048min: 1, max: 65536
presence_penaltynumber

Presence penalty. Positive values penalize tokens that have already appeared, encouraging the model to talk about new topics.

Default: 0min: -2, max: 2
promptstring

The prompt to send to the model.

Default: ""
system_promptstring

System prompt to guide the model's behavior.

Default: "You are a helpful assistant."
temperaturenumber

The value used to modulate the next token probabilities.

Default: 0.7min: 0, max: 2
top_pnumber

A probability threshold for nucleus sampling. Only keeps the top tokens with cumulative probability >= top_p.

Default: 0.8min: 0, max: 1
Version: b964682d1c19Updated: 7/25/202614 runs