The Open-Source AI Stack
RSS
All models

Models · qwen

QwQ 32B Preview

Open Alibaba · 2024-11-27 · Apache-2.0

Alibaba's first public reasoning model, fine-tuned from Qwen 2.5 32B Instruct and shipped as a research preview a month after OpenAI's o1-preview. Apache 2.0 under the Qwen umbrella, with the explicit framing of contemplative problem-solving rather than fast token generation.

Cost

— / Mtok input
— / Mtok output

Together AI · as of 2026-05-19

via Artificial Analysis ↗

Architecture

tokens in Embedding vocab not disclosed · qwen tokenizer × 64 layers Grouped-Query Attention RoPE context 32,768 tokens Dense MLP SwiGLU activation (standard) 32.5B active params Output projection tokens out
Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture
dense
Total params
32.5B
Active params
32.5B
Context window
33K tokens
Attention
gqa
Position encoding
rope
Post-training
sft
OSI-approved
yes
Data released
no
Training code
not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

Code

LiveCodeBench 50.0 as of 2024-11-28 source ↗

Math

AIME 2024 50.0 as of 2024-11-28 source ↗

Available quantizations

GGUF llama.cpp's container; the common local format, k-quants from Q2 to Q8. runs on llama.cpp, Ollama
AWQ Activation-aware 4-bit weight quantization for GPU serving. runs on vLLM, SGLang
GPTQ Post-training 4-bit weight quantization for GPU serving. runs on vLLM, SGLang, Transformers
EXL2 ExLlamaV2's variable-bitrate format for consumer GPUs. runs on ExLlamaV2
MLX Apple MLX 4/8-bit layout for Apple silicon. runs on Apple MLX
FP8 8-bit float, frequently a native release on Hopper / Blackwell GPUs. runs on vLLM, SGLang, TensorRT-LLM
bitsandbytes On-the-fly NF4 / INT8 weight quantization inside Transformers. runs on Transformers

Verified via the Hugging Face model tree ↗. Community quantizations change over time; the families shown are those with published weights at audit time.

Notable innovations

  • · First open-weights o1-style reasoning preview at the 32B scale
  • · Apache 2.0 reasoning model
  • · Built on Qwen 2.5 32B Instruct base

Sources