72B dense base model
The model card identifies this repository as the 72B Qwen2 base language model, matching about 72.71B captured parameters.
Open Source Model Profile · Qwen
Qwen2-72B
Qwen2-72B is a Qwen 72.71B base text-generation model. Its card documents a dense Transformer design and post-training guidance.
- Publisher
- Qwen
- Task
- text-generation
- Model type
- qwen2
- License
- tongyi-qianwen
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Qwen2-72B is published by Qwen as a text-generation base model. The captured configuration identifies Qwen2ForCausalLM with Safetensors metadata reporting 72,706,203,648 parameters. According to the model card, this repository holds the 72B Qwen2 base language model in a series ranging from 0.5 to 72 billion parameters.
Recorded capabilities
SwiGLU, QKV bias, and GQA design
According to the model card, the series uses SwiGLU activation, attention QKV bias, group query attention, and an improved tokenizer for multiple natural languages and code.
Post-training guidance
The card advises against direct text generation with the base model and recommends SFT, RLHF, or continued pretraining.
Broad publisher evaluation list
According to the model card, documented suites span language understanding, coding, mathematics, Chinese, and multilingual tasks with a multi-model comparison table.
Use cases in the source record
- Post-training research with SFT, RLHF, or continued pretraining on top of the documented 72B base weights.
- Multilingual, coding, and mathematics evaluation experiments that follow the card's documented benchmark suites.
Limitations and unknowns
- No independently extracted benchmark scores were included; comparison-table claims are publisher material.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data and should be refreshed before being presented as current availability.
- Competitiveness characterizations and benchmark comparisons come from the publisher model card and have not been independently verified by Ethen.
Source and provenance
Source: Qwen/Qwen2-72B
Captured: Unknown. Processed: 2026-09-07T19:34:35.754543+00:00.
Qwen2-72B Introduction Qwen2 is the new series of Qwen large language models. For Qwen2, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters, including a Mixture-of-Experts model. This repo contains the 72B Qwen2 base language model. Compared with the state-of-the-art opensource language models, including the previous released Qwen1.5, Qwen2 has generally surpassed most opensource models and demonstrated competitiveness against proprietary models across a series of benchmarks targeting for language understanding, language generation, multilingual capability, cod…
F001F002F003F004F005F006F007F008F009F010F011F012F014F015F016