LANGUAGE MODEL Alibaba Last updated:

QwQ-32B Preview

Alibaba's Open Reasoning Model

Alibaba's first open-weight reasoning model, released November 2024 as a "preview." Used the same RL-on-reasoning-traces recipe OpenAI used for o1, but with the model weights downloadable — making it the first time anyone outside the frontier US labs could study how reasoning capability is trained in. Its release 8 weeks before DeepSeek R1 partly set the expectation that "open reasoning" would arrive within months, not years.

Intelligence
Below avg
Cost
Free
Open weights — self-host
Context
33K
Up to 32,768 tokens
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on Artificial Analysis. Cost buckets are fixed dollar thresholds keyed off output-token price ($/M out).
Intelligence
  • Top 1%≤ 1%
  • Top 5%≤ 5%
  • Top 10%≤ 10%
  • Good≤ 25%
  • Medium≤ 50%
  • Below avg> 50%
Speed
  • Top 1%≥ 345 tok/s
  • Top 5%≥ 237 tok/s
  • Top 10%≥ 196 tok/s
  • Good≥ 146 tok/s
  • Medium≥ 90 tok/s
  • Slow< 90 tok/s
Cost
  • Freeopen weights · self-host
  • Low< $1 / M out
  • Moderate$1–5 / M out
  • High≥ $5 / M out

Why it matters

QwQ proved that the o1 recipe was reproducible from public research alone — no OpenAI insider knowledge required. That proof, more than the model's specific capabilities, was the contribution. By Q1 2025, ten different labs had released reasoning models; the technique had become a shared methodology.

Core Capabilities

Long Documents
Handles entire codebases, books, and multi-doc RAG.
Generative
Produces images, video, audio, or other media.
Agent Workflows
Built for tool use and autonomous tasks.

Context Window

33k tokens
≈ 25 pages
4k Chat 聊天
32k This model 本模型
128k Books 整本书
400k Multi-doc 多文档
1M Codebase 整个代码库
10M

Availability

API
Not available
Product / App
Not available
Open Source
Released
Enterprise

Pricing Model

Free / self-host
Open weights — pay only for compute.
Self-host

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model
Reasoning
AA Intelligence Index · scaled to 10
1.7
5.6
2.2
Coding
SciCode · scaled to 10
1.8
4.3
0.4
Context / memory
Context window size · log-scaled
6.0
9.0
4.0
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

  • Language model from Alibaba — see the linked sources below for benchmark and review coverage
  • Tool-use and agent loops are the typical fit per the published model card

Best use cases

  • Agent / tool-use workflows that match the model's published benchmarks
  • See the model spec and sources block for benchmarked use cases

Tools to try

Not ideal for

  • Tasks far outside the modalities listed in this model's spec
  • Workflows where a more recent successor in the same family scores higher