131K-token context
According to the model card, the model supports a full 131,072-token context length.
Open Source Model Profile · fluently
FluentlyQwen2.5-32B
FluentlyQwen2.5-32B is a 32.5B-parameter Qwen2 causal language model from fluently, also called FluentlyLM Prinum. Its model card documents 131,072-token context, seven-language support, and MIT licensing.
- Publisher
- fluently
- Task
- text-generation
- Model type
- qwen2
- License
- mit
- Library
- transformers
- Publication status
- Approved for indexing
Model overview
FluentlyQwen2.5-32B is published by fluently as a Qwen2 text-generation model, also called FluentlyLM Prinum in its 32B version. The captured configuration identifies Qwen2ForCausalLM with model type qwen2, and Safetensors metadata reports 32,763,876,352 parameters against the model card's stated 32.5B. The model card documents 131,072-token context, 64 layers with grouped-query attention, seven-language support, and MIT licensing.
Recorded capabilities
Seven-language coverage
According to the model card, English, French, Spanish, Russian, Chinese, Japanese, and Persian carry official support.
64-layer GQA architecture
According to the model card, the model has 64 layers with grouped-query attention using 40 Q heads and 8 KV heads.
Reasoning and instruct tagging
Hub tags mark instruct, math, roleplay, reasoning, and code themes alongside fluently-lm and dataset entries for ultraset, ultrathink, reasoning-1-1k, and MATH-500-Overall.
MIT licensing
Card data, Hub tags, and the model card all record the MIT license.
Use cases in the source record
- Multilingual instruction-following and chat generation across the seven officially supported languages.
- Math, reasoning, roleplay, and code generation consistent with the instruct, math, roleplay, reasoning, and code tagging and the model card's code example.
Limitations and unknowns
- The model card embeds metric figures without a clear evaluation table, so no independently verified evaluation results are stated.
- No training epochs, hardware, or duration were extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: fluently/FluentlyQwen2.5-32B
Captured: Unknown. Processed: 2026-09-07T19:35:21.632209+00:00.
FluentlyQwen2.5 32B (a.k.a FluentlyLM Prinum (32B-version) Introducing the first standalone model from Project Fluently LM! We worked on it for several months, used different approaches, and eventually found the optimal one. Model Details Model Description Developed by: @fluently-lm Model type: Causal Language Models (QwenForCausalLM, LM Transformer) Number of Parameters: 32.5B Number of Paramaters (Non-Embedding): 31.0B Number of Layers: 64 Number of Attention Heads (GQA): 40 for Q and 8 for KV Context Length: Full 131,072 tokens Language(s) (NLP): English, French, Spanish, Russian, Chinese, Japanese, Persian (official support) Lic…
F001F002F003F004F005F006F007F009F010F011F012F015F017