Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features: - Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios. - Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and…
Open-weight model · Text generation
Qwen3-32B-german-city-names-second-third-sft-bf16
by Localized Ft localized-ft/Qwen3-32B-german-city-names-second-third-sft-bf16
Qwen3-32B-german-city-names-second-third-sft-bf16 is an open-weight model for text generation from Localized Ft, released under Apache License 2.0. It has 32.8B parameters and a 40,960-token context. At 16-bit it needs about 78.6 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.
Parameters32.8B
Context40,960
Weights65.5 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve Qwen3-32B-german-city-names-second-third-sft-bf16 (32.8B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 65.5 GB | 78.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 32.8 GB | 39.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 16.4 GB | 19.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 9, 2026.
Model Card
By Localized Ft, published under apache-2.0, revision 7db35ebba497.
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.
Read Localized Ft's full model card
Uploaded finetuned model
- Developed by: localized-ft
- License: apache-2.0
- Finetuned from model : unsloth/Qwen3-32B
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 40,960
- Layers
- 64
- Hidden size
- 5,120
- Feed-forward size
- 25,600
- Attention heads
- 64
- Key/value heads
- 8
- Head dimension
- 128
- Vocabulary size
- 151,936
- Stored precision
- bfloat16
- Model type
- qwen3
Identity and Version
- Repository
- localized-ft/Qwen3-32B-german-city-names-second-third-sft-bf16
- Publisher
- Localized Ft
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 32.8B parameters
- Languages
- en
- Revision
- 7db35ebba497e4bd1882c638ff4943ae9f63eaa0
- First published
- 2026-09-27
- Last updated
- 2026-09-27
Files and Weights
21 files, 65.5 GB in total. The weights are 14 files totalling 65.5 GB in safetensors.
Weights14 files · 65.5 GB
Configuration2 files · 60.9 KB
Tokenizer2 files · 11.4 MB
Documentation1 file · 563 B
Other1 file · 4.8 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00014.safetensors | Weights | 4.9 GB | e5da875a02c8 |
| model-00002-of-00014.safetensors | Weights | 4.9 GB | 007c079c15ca |
| model-00003-of-00014.safetensors | Weights | 4.9 GB | 56e2b3a80f93 |
| model-00004-of-00014.safetensors | Weights | 4.9 GB | c8eceb24e1e8 |
| model-00005-of-00014.safetensors | Weights | 4.9 GB | 7c3e39c6441f |
| model-00006-of-00014.safetensors | Weights | 4.9 GB | 4188854886ba |
| model-00007-of-00014.safetensors | Weights | 4.9 GB | c7b93689a1db |
| model-00008-of-00014.safetensors | Weights | 4.9 GB | d3ef26ecb3ed |
| model-00009-of-00014.safetensors | Weights | 4.9 GB | 227c9afd2303 |
| model-00010-of-00014.safetensors | Weights | 4.9 GB | c2cfea0d51e4 |
| model-00011-of-00014.safetensors | Weights | 4.9 GB | 397f63feb7cf |
| model-00012-of-00014.safetensors | Weights | 4.9 GB | 1dad948134cc |
| model-00013-of-00014.safetensors | Weights | 4.9 GB | a504a2fa85eb |
| model-00014-of-00014.safetensors | Weights | 2.1 GB | ae5c47ee8b51 |
| config.json | Configuration | 2.6 KB | — |
| model.safetensors.index.json | Configuration | 58.3 KB | — |
| README.md | Documentation | 563 B | — |
| chat_template.jinja | Other | 4.8 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 11.4 MB | ca634ec77666 |
| tokenizer_config.json | Tokenizer | 5.3 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 65.5 GB
Download from Localized Ft
Released by Localized Ft through its official repository on Hugging Face. Read the license.
Built From
- Derived from unsloth/Qwen3-32B
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 65.5 GB |
| 16-bit | 65.5 GB |
| 8-bit | 32.8 GB |
| 4-bit | 16.4 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Qwen3-32B-german-city-names-second-third-sft-bf16
How much GPU memory does Qwen3-32B-german-city-names-second-third-sft-bf16 need?
About 78.6 GB at 16-bit and 19.7 GB at 4-bit: the weights (32.8B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Qwen3-32B-german-city-names-second-third-sft-bf16 on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Qwen3-32B-german-city-names-second-third-sft-bf16 commercially?
Yes. Qwen3-32B-german-city-names-second-third-sft-bf16 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
What is Qwen3-32B-german-city-names-second-third-sft-bf16's context length?
40,960 tokens, from the maximum position embeddings in its published configuration.
Similar Models
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features: - Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios. - Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and…
MXFP4 W4A4 quantization of (revision 9216db5781bf21249d130ec9da846c4624c16137, BF16). It was made for a controlled NVFP4-vs-MXFP4 decode benchmark on NVIDIA B200 (fp4bench), not as a general-purpose release: the NVFP4 and MXFP4 checkpoints share the model, the tool, the recipe, the calibration data and the layer coverage, and differ only in the format. Recipe: QuantizationModifier(targets="Linear", scheme="MXFP4", ignore=["lmhead"]). The plain preset recipe, with no weight-rounding optimization (GPTQ, AutoRound and the like). anon8231489123/ShareGPTVicunaunfiltered at revision 192ab2185289094fc556ec8ce5ce1e8e587154ca, at most 1024 tokens each (33208 tokens in all, seed 3). MXFP4 has no…
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.
This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library.