Code-specific Qwen series
According to the model card, Qwen2.5-Coder covers six sizes from 0.5B to 32B and improves code generation, reasoning, and fixing over CodeQwen1.5.
Open Source Model Profile · Qwen
Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is a 32.76B-parameter Qwen2 text-generation model from Qwen. According to the model card, it is a code-specific model with long-context support.
- Publisher
- Qwen
- Task
- text-generation
- Model type
- qwen2
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Qwen2.5-Coder-32B is published by Qwen as a Transformers text-generation model. The captured configuration identifies Qwen2ForCausalLM with model type qwen2 and about 32.76B Safetensors parameters under Apache-2.0. According to the model card, it is the 32B member of the code-specific Qwen2.5-Coder series.
Recorded capabilities
32B-class Qwen2 build
Captured configuration identifies Qwen2ForCausalLM with 32763876352 Safetensors parameters; the card states 32.5B total with 31.0B non-embedding parameters.
Long-context support
According to the model card, the model supports up to 128K tokens, with full stated context of 131,072 tokens and deployment notes for long texts.
Apache-2.0 code model
Captured card data records apache-2.0, and hub tags mark code, qwen-coder, conversational, and endpoints-compatible use.
Post-training starting point
According to the model card, users can apply SFT, RLHF, continued pretraining, or fill-in-the-middle work, with vLLM documentation referenced for deployment.
Use cases in the source record
- Code generation, reasoning, and fixing workflows using the publisher's code-specific Qwen2.5-Coder setup.
- Long-context code and Code Agent experiments using the reported 128K-token support and vLLM deployment notes.
Limitations and unknowns
- No Ethen-measured evaluation results were extracted; capability comparisons are publisher-reported model-card claims.
- GPU memory and throughput requirements are referenced only as an external link in the model card and were not extracted.
- No pricing or current provider availability was extracted; provider state is historical snapshot data.
Source and provenance
Source: Qwen/Qwen2.5-Coder-32B
Captured: Unknown. Processed: 2026-09-07T19:34:35.992498+00:00.
Qwen2.5-Coder-32B Introduction Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). As of now, Qwen2.5-Coder has covered six mainstream model sizes, 0.5, 1.5, 3, 7, 14, 32 billion parameters, to meet the needs of different developers. Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: Significantly improvements in code generation , code reasoning and code fixing . Base on the strong Qwen2.5, we scale up the training tokens into 5.5 trillion including source code, text-code grounding, Synthetic data, etc. Qwen2.5-Coder-32B has become the current state-of-the-art o…
F001F002F003F004F005F006F007F009F010F011F012F013F014F015F016F017