0.5B code-specific base model
According to the model card, this is the 0.5B Qwen2.5-Coder causal language model at the pretraining stage, within a series covering 0.5B to 32B parameters.
Open Source Model Profile · Qwen
Qwen2.5-Coder-0.5B
Qwen2.5-Coder-0.5B is a 0.49B-parameter Qwen2 code-generation model from Qwen at the pretraining stage, with a documented 32,768-token context.
- Publisher
- Qwen
- Task
- text-generation
- Model type
- qwen2
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Qwen2.5-Coder-0.5B is published by Qwen as a code-specific Qwen2 text-generation model. The captured configuration identifies Qwen2ForCausalLM and Safetensors metadata reports 494,032,768 parameters. According to the model card, this repo holds the 0.5B pretraining-stage model with 32,768 tokens of context, and card data records apache-2.0.
Recorded capabilities
32K documented context
According to the model card, the model supports 32,768 tokens of context with grouped-query attention of 14 query heads and 2 key-value heads.
Code generation and reasoning focus
According to the model card, the series targets code generation, code reasoning, and code fixing, with a stated foundation for code-agent applications.
Memory and throughput reference
According to the model card, GPU-memory requirements and throughput are documented through an external results link rather than in the extracted record.
Use cases in the source record
- Code generation, code reasoning, and code-fixing experiments using the publisher's documented pretraining-stage setup.
- Post-training and fill-in-the-middle development, which the model card suggests instead of direct conversational use.
Limitations and unknowns
- No evaluation results were extracted from this record.
- GPU-memory and throughput figures are referenced only as an external link in the model card and were not extracted.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: Qwen/Qwen2.5-Coder-0.5B
Captured: Unknown. Processed: 2026-09-07T19:34:35.863296+00:00.
Qwen2.5-Coder-0.5B 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…
F001F002F003F004F005F006F007F009F010F011F012F013F014F015