OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for Singapore / International. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.04 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Explicit cache · Input<=32k
Explicit cache write1
$0.25
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.02
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.375
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.03
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.75
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.06
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for China (Beijing). Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.029 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Batch · Input<=32k
Input(Batch File)1
$0.072
1M tokens
Batch · Input<=32k
Output(Batch File)1
$0.717
1M tokens
Explicit cache · Input<=32k
Explicit cache write1
$0.18
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.014
1M tokens
Batch · Input<=32k
Input(Batch Chat)1
$0.144
1M tokens
Batch · Input<=32k
Output(Batch Chat)1
$1.434
1M tokens
Batch · 32k<Input<=128k
Input(Batch File)2
$0.108
1M tokens
Batch · 32k<Input<=128k
Output(Batch File)2
$1.075
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.27
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.022
1M tokens
Batch · 32k<Input<=128k
Input(Batch Chat)2
$0.216
1M tokens
Batch · 32k<Input<=128k
Output(Batch Chat)2
$2.151
1M tokens
Batch · 128k<Input<=256k
Input(Batch File)3
$0.215
1M tokens
Batch · 128k<Input<=256k
Output(Batch File)3
$2.15
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.539
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.043
1M tokens
Batch · 128k<Input<=256k
Input(Batch Chat)3
$0.431
1M tokens
Batch · 128k<Input<=256k
Output(Batch Chat)3
$4.301
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for Germany (Frankfurt) / EU. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.04 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Explicit cache · Input<=32k
Explicit cache write1
$0.25
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.02
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.375
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.03
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.75
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.06
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for Germany (Frankfurt) / Global. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.029 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Explicit cache · Input<=32k
Explicit cache write1
$0.18
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.014
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.27
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.022
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.539
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.043
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for US (Virginia) / Global. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.029 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Explicit cache · Input<=32k
Explicit cache write1
$0.18
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.014
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.27
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.022
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.539
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.043
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
OpenAI-compatible Chat Completions or DashScope, with a regional workspace API key. Listed scope describes this price/capability configuration; account eligibility must be checked.
USD list rates for Hong Kong (China) / Hong Kong. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion. Cached input price: $0.04 / 1M tokens (base tier).
Published billing units · original currency · observed 12 Oct 2026
Category
Usage
Rate
Unit
Explicit cache · Input<=32k
Explicit cache write1
$0.25
1M tokens
Explicit cache · Input<=32k
Explicit cache read1
$0.02
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache write2
$0.38
1M tokens
Explicit cache · 32k<Input<=128k
Explicit cache read2
$0.03
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache write3
$0.75
1M tokens
Explicit cache · 128k<Input<=256k
Explicit cache read3
$0.06
1M tokens
Rate conditions
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
USD list rates for Singapore / International. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
USD list rates for China (Beijing). Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
USD list rates for Germany (Frankfurt) / EU. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
Caching reviewed 12 Oct 2026
Alibaba Model Studio · Germany (Frankfurt) · Global
USD list rates for Germany (Frankfurt) / Global. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
USD list rates for US (Virginia) / Global. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
Caching reviewed 12 Oct 2026
Alibaba Model Studio · Hong Kong (China) · Hong Kong
USD list rates for Hong Kong (China) / Hong Kong. Cached input means implicit cache reads. The total input length selects the rate for the whole request. K=1,000; prices beyond the final listed tier are not verified. Separate explicit-cache and batch charges are listed below. Published card prices are retained without currency conversion.
Input tier: Input<=32k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 32k<Input<=128k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Input tier: 128k<Input<=256k. Alternative billing mode for this same endpoint. Explicit creation replaces normal input billing for written tokens; explicit reads replace input billing for matching tokens. Batch prices are separate and do not stack with cache discounts.
Observed 12 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
Caching reviewed 12 Oct 2026
What the measurements say
No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model
Why it is included
The provider explicitly distinguishes its Qwen3-VL closed-source series from named open-weight Qwen3-VL checkpoints. This record is the Plus service, not an open-weight size variant.
