The Open-Source AI Stack
RSS
All models

Models · qwen

Qwen 2.5 Coder 32B Instruct

Open Alibaba · 2024-11-12 · Apache-2.0

Code-specialized 32B sibling to Qwen 2.5, released alongside 0.5B / 1.5B / 7B / 14B / 32B coder variants all under Apache 2.0. Alibaba positioned it as comparable to GPT-4o on Aider code-repair and first among open models on multi-language repair.

Cost

$0.00 / Mtok input
$0.00 / Mtok output

Together AI · as of 2026-05-21

via Artificial Analysis ↗

Speed

0 tok/sec output
0 ms TTFT

· as of 2026-05-21

source ↗

Architecture

tokens in Embedding vocab not disclosed · qwen tokenizer × 64 layers Grouped-Query Attention RoPE context 131,072 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
131K tokens
Attention
gqa
Position encoding
rope
Post-training
sft, dpo
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.

General reasoning

MMLU-Pro 63.5 as of 2026-05-21 source ↗
GPQA-Diamond 41.7 as of 2026-05-21 source ↗

Code

LiveCodeBench 29.5 as of 2026-05-21 source ↗

Math

MATH 76.7 as of 2026-05-21 source ↗
AIME 2024 12.0 as of 2026-05-21 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

  • · 40+ programming language coverage
  • · Fill-in-the-Middle SOTA across 5 benchmarks
  • · Apache 2.0 across full coder size ladder

Sources