Skip to main content
Allgate
Canvas Start free
MODEL LIBRARY

Alibaba (China)alibaba-cn / qwen-flash

Qwen Flash

Efficient Qwen model for fast chat, extraction, and high-volume workloads

Provider
Alibaba (China) (alibaba-cn)
Model identifier
qwen-flash
Library category
Chat
Source update
2025-07-28
Release date
2025-07-28
Provider documentation
https://www.alibabacloud.com/help/en/model-studio/models

Source pricing and conditions

Costs retain the source structure, tiers and conditions. These snapshot values are not Allgate API quotes. Cost currency and provider billing currency are separate fields; each defaults to USD when absent.

Cost currency: USD · Provider billing currency: CNY. Input, output, reasoning, cache and audio token fields use the cost currency per million tokens. Explicit per-image, per-second and per-10,000-character fields retain their named units and resolution conditions.

Tier prices apply above the stated size. When tiers exist, source logic ignores the legacy context_over_200k field. Scheduled prices override stated fields within their window; omitted fields inherit base prices. Source pricing schema

Cost currency differs from provider billing currency. The source may contain upstream estimates; no currency conversion has been applied here.

  • pricing: "alibaba-cn" bills in CNY but 53 model(s) still carry upstream USD estimates; add cost overrides with "currency": "CNY"
cache_read
0.0044
cache_write
0.0275
input
0.022
output
0.216
tiers
  1. cache_read
    0.0174
    cache_write
    0.10875
    input
    0.087
    output
    0.861
    tier
    size
    128000
    type
    context
  2. cache_read
    0.0346
    cache_write
    0.21625
    input
    0.173
    output
    1.721
    tier
    size
    256000
    type
    context

Model metadata

Original source field names and values are preserved. False, zero, empty objects and missing values remain distinct. Missing capabilities are not inferred from a name, category or related version.

attachment
false
description
Efficient Qwen model for fast chat, extraction, and high-volume workloads
family
qwen
id
qwen-flash
knowledge
2024-04
last_updated
2025-07-28
limit
context
1000000
output
32768
modalities
input
  1. text
output
  1. text
name
Qwen Flash
open_weights
false
reasoning
true
reasoning_options
  1. type
    toggle
  2. max
    81920
    type
    budget_tokens
release_date
2025-07-28
structured_output
true
temperature
true
tool_call
true

Provider reference

Provider documentation and billing metadata belong to the listed source provider. They do not describe Allgate routing or account configuration.

id
alibaba-cn
name
Alibaba (China)
doc
https://www.alibabacloud.com/help/en/model-studio/models
currency
CNY
billing
priority
20
thinkingToggle
param
enable_thinking
value
true

About this reference

This library reproduces the public llm-metadata snapshot: 592 model entries from 50 providers. This community maintained source does not independently verify every provider claim. Model identifiers belong to their listed provider; matching names from different providers remain separate entries. llm-metadata

Reference metadata does not establish Allgate availability or prices. Check the live model catalogue for current service and rates. Missing fields remain unknown. Source update dates are not substituted for release dates; retrieval dates are not model update dates. View live catalogue

Source: llm-metadata · Apache-2.0 · Retrieved · Source generated · Snapshot provenance