Skip to navigation

Qwen3.6 Flash

Qwen3.6 Flash
Alibaba Cloud · Text Generation
POST /v1/chat/completions

Fast Qwen3.6 vision-language model for agentic coding, math reasoning, spatial understanding, OCR, and text, image, and video input.

See Qwen3.6 Flash pricing and specs or try it in the Playground.

At a glance

FieldValue
Model idqwen3-6-flash
Model release date2026-04-16
Input modalitiesText, Image, Video
Output modalitiesText
Context window983.6K
Weight precision-
Max output tokens65,536
RegionSingapore
Featuresreasoning, vision, video, web_search, function_calling, agentic_coding
Native inferenceNo
NewNo
Structured outputJSON Mode
Supported endpointsPOST /v1/chat/completions, POST /v1/responses, POST /v1/messages, POST /v1beta/models/qwen3-6-flash:generateContent
Alternate model idsqwen3.6-flash

Pricing

ChargeSpecRate
Inputper 1M prompt tokens<=256K $0.25; 256K-1M $1.00
Outputper 1M generated tokens<=256K $1.50; 256K-1M $4.00
Web searchper request when enabled$0.02

Example request

curl https://api.empiriolabs.ai/v1/chat/completions \
-H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
-H 'Content-Type: application/json' \
-d '{"model": "qwen3-6-flash", "messages": [{"role":"user","content":"Hello"}]}'

Parameters

ParameterTypeRequiredDefaultDescription
temperaturenumberno0.7Sampling temperature. 0 is deterministic and 1.99 is maximum randomness. · Range: 0 – 1.99
top_pnumberno0.9Nucleus sampling probability mass. Lower values make outputs more focused. · Range: 0 – 1
max_tokensnumberno-Maximum output tokens. When omitted, the model’s own default applies. · Range: 1 – 65536
stopstringno-Up to 4 strings where the model will stop generating further tokens.
enable_thinkingbooleannotrueEnable reasoning before answering. With response_format, the model answers without reasoning unless the request sets enable_thinking. A request that forces a tool call runs without reasoning unless it sets enable_thinking.
reasoning_effortenumno"medium"Reasoning effort level. none disables thinking. low, medium, high, and max set bounded thinking budgets sized to the selected model. · Allowed: none, low, medium, high, max
thinking_budgetnumberno32768Maximum tokens reserved for reasoning when thinking is enabled. · Range: 1 – 64000
preserve_thinkingbooleannofalseInclude the reasoning of earlier assistant turns in the model’s input on /v1/chat/completions. Return each turn’s reasoning unchanged in that assistant message’s reasoning_content field. The earlier reasoning is billed as input tokens.
seedintegerno-Random seed for more reproducible sampling. Reusing a seed with the same request returns similar, not always identical, output. · Range: 0 – 2147483647
top_kintegerno-Sample only from the K most likely tokens. When omitted, the model’s own default applies. · Range: 1 – 100
presence_penaltynumberno-Penalty for tokens that already appeared in the generated text. Positive values encourage new topics. When omitted, the model’s own default applies. · Range: -2 – 2
repetition_penaltynumberno-Higher values discourage repeated phrasing, and 1 applies no penalty. When omitted, the model’s own default applies. · Range: 0.1 – 2
nintegerno1Number of answers to return. Above 1 runs only on /v1/chat/completions, with enable_thinking false and without tools, web search or logprobs. The output tokens of every answer are billed. · Range: 1 – 4
logprobsbooleannofalseReturn the log probabilities of the answer tokens, not the reasoning, on /v1/chat/completions.
top_logprobsintegerno0Return up to this many most likely alternatives for each answer token. Needs logprobs set to true. · Range: 0 – 5
vl_high_resolution_imagesbooleanno-Use higher resolution processing for image inputs. When omitted, the model’s own default applies.
max_pixelsnumberno2621440Maximum pixel count per image when high resolution processing is disabled. · Range: 4096 – 16777216
video_fpsnumberno2Frames per second to sample from video inputs. · Range: 0.1 – 10
tool_web_searchbooleannofalseSearch the web for real-time information. Adds $0.02 to the request cost when enabled.
parallel_tool_callsbooleanno-Allow multiple tool calls in a single assistant turn. When omitted, the model’s own default applies.
response_formatenumno-Return the output as a valid JSON object (JSON mode). Describe the fields you want in your prompt.

Notes

Supports text, image, and video input. Web search is available through tool_web_search and adds $0.02 per request when enabled. Thinking tokens are billed as output tokens. Explicit cache controls are not supported.

Variants

:variant1

FieldValue
Model idqwen3-6-flash:variant1
Model release date2026-04-16
RegionChina
Context window983.6K
Weight precision-
Max output tokens65,536
Featuresreasoning, vision, video, web_search, function_calling, agentic_coding
Native inferenceNo
Structured outputJSON Mode
Batch API35% off list price
Supported endpointsPOST /v1/chat/completions, POST /v1/responses, POST /v1/messages, POST /v1beta/models/qwen3-6-flash:variant1:generateContent

Pricing

ChargeSpecRate
Inputper 1M prompt tokens<=256K $0.25 $0.165; 256K-1M $1.00 $0.66
Outputper 1M generated tokens<=256K $1.50 $0.99; 256K-1M $4.00 $3.961
Web searchper request when enabled$0.01

Parameters

ParameterTypeRequiredDefaultDescription
temperaturenumberno0.7Sampling temperature. 0 is deterministic and 1.99 is maximum randomness. · Range: 0 – 1.99
top_pnumberno0.9Nucleus sampling probability mass. Lower values make outputs more focused. · Range: 0 – 1
max_tokensnumberno-Maximum output tokens. When omitted, the model’s own default applies. · Range: 1 – 65536
stopstringno-Up to 4 strings where the model will stop generating further tokens.
enable_thinkingbooleannotrueEnable reasoning before answering. With response_format, the model answers without reasoning unless the request sets enable_thinking. A request that forces a tool call runs without reasoning unless it sets enable_thinking.
reasoning_effortenumno"medium"Reasoning effort level. none disables thinking. low, medium, high, and max set bounded thinking budgets sized to the selected model. · Allowed: none, low, medium, high, max
thinking_budgetnumberno32768Maximum tokens reserved for reasoning when thinking is enabled. · Range: 1 – 128000
preserve_thinkingbooleannofalseInclude the reasoning of earlier assistant turns in the model’s input on /v1/chat/completions. Return each turn’s reasoning unchanged in that assistant message’s reasoning_content field. The earlier reasoning is billed as input tokens.
seedintegerno-Random seed for more reproducible sampling. Reusing a seed with the same request returns similar, not always identical, output. · Range: 0 – 2147483647
top_kintegerno-Sample only from the K most likely tokens. When omitted, the model’s own default applies. · Range: 1 – 100
presence_penaltynumberno-Penalty for tokens that already appeared in the generated text. Positive values encourage new topics. When omitted, the model’s own default applies. · Range: -2 – 2
repetition_penaltynumberno-Higher values discourage repeated phrasing, and 1 applies no penalty. When omitted, the model’s own default applies. · Range: 0.1 – 2
nintegerno1Number of answers to return. Above 1 runs only on /v1/chat/completions, with enable_thinking false and without tools, web search or logprobs. The output tokens of every answer are billed. · Range: 1 – 4
logprobsbooleannofalseReturn the log probabilities of the answer tokens, not the reasoning, on /v1/chat/completions.
top_logprobsintegerno0Return up to this many most likely alternatives for each answer token. Needs logprobs set to true. · Range: 0 – 5
vl_high_resolution_imagesbooleanno-Use higher resolution processing for image inputs. When omitted, the model’s own default applies.
max_pixelsnumberno2621440Maximum pixel count per image when high resolution processing is disabled. · Range: 4096 – 16777216
video_fpsnumberno2Frames per second to sample from video inputs. · Range: 0.1 – 10
tool_web_searchbooleannofalseSearch the web for real-time information. Adds $0.01 to the request cost when enabled.
parallel_tool_callsbooleanno-Allow multiple tool calls in a single assistant turn. When omitted, the model’s own default applies.
response_formatenumno-Return the output as a valid JSON object (JSON mode). Describe the fields you want in your prompt.

Notes

Supports text, image, and video input. Web search is available through tool_web_search and adds $0.01 per request when enabled. Thinking tokens are billed as output tokens. Explicit cache controls are not supported.


Machine-readable schema: GET https://api.empiriolabs.ai/v1/models/qwen3-6-flash.