This model is a fine-tuned version of None. It has been trained using TRL. This model was trained with SFT.
FreeSolo-B-341M is an open-weight model for text generation from Zekun Shi, released under MIT License. It has 374M parameters. At 16-bit it needs about 0.9 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
RoPE-free long-range attention, trained at 4k with zero-shot length extrapolation. FS-B searches full-prefix and norm-filtered forests of chunk mean keys, reranks their candidates, and reads four chunks. Its attention is superlinear and subquadratic.
Parameters374M
Context—
Weights1.5 GB
Licensemit
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve FreeSolo-B-341M (374M 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 | 0.7 GB | 0.9 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.4 GB | 0.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.2 GB | 0.2 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 16, 2026.
FreeSolo-B-341M on every accelerator the SAVRN Index prices, at every precision
Model Card
By Zekun Shi, published under mit, revision 340f7219f163.
RoPE-free long-range attention, trained at 4k with zero-shot length extrapolation. FS-B searches full-prefix and norm-filtered forests of chunk mean keys, reranks their candidates, and reads four chunks. Its attention is superlinear and subquadratic. This 341M research base model was trained on FineWeb-Edu at a 4,096-token context. The short retrieval example asks for the hidden number 5437923. Start with CUDA-enabled PyTorch; Usage gives the tested installation and speed-replication command. FreeSolo learns simple chunk mean keys that support search, norm pruning, and hierarchical summarization. RoPE stays in the local window; long-range attention reads selected original tokens without…
Read Zekun Shi's full model card
Paper · GitHub · Models · Usage
RoPE-free long-range attention, trained at 4k with zero-shot length extrapolation.
FS-B searches full-prefix and norm-filtered forests of chunk mean keys, reranks their candidates, and reads four chunks. Its attention is superlinear and subquadratic. This 341M research base model was trained on FineWeb-Edu at a 4,096-token context.
Try it
hf download zekun-shi/FreeSolo-B-341M freesolo-source.zip --local-dir ./FreeSolo-B-341M
python -m pip install ./FreeSolo-B-341M/freesolo-source.zip
freesolo-generate --model zekun-shi/FreeSolo-B-341M --output generation.json
The short retrieval example asks for the hidden number 5437923. Start with CUDA-enabled PyTorch; Usage gives the tested installation and speed-replication command.
The idea
FreeSolo learns simple chunk mean keys that support search, norm pruning, and hierarchical summarization. RoPE stays in the local window; long-range attention reads selected original tokens without rotating their queries and keys.
Results
This is the 341M checkpoint used for the paper's FS-B speed measurements. See speed replication for the complete protocol and expected throughput.
This is a base model without instruction tuning. Retrieval is task dependent, especially when several facts or values must be combined. Training details, artifact checksums, and citation are in Usage.
Configuration
- Vocabulary size
- 32,000
- Model type
- freesolo
Identity and Version
- Repository
- zekun-shi/FreeSolo-B-341M
- Publisher
- Zekun Shi
- Task
- Text generation
- Modality
- Text
- Library
- pytorch
- Parameters
- 374M parameters
- Languages
- en
- Revision
- 340f7219f163ba98a4538815a682469d098eabf0
- First published
- 2026-10-15
- Last updated
- 2026-10-16
Files and Weights
11 files, 1.5 GB in total. The weights are 1 file totalling 1.5 GB in safetensors.
Weights1 file · 1.5 GB
Configuration2 files · 12.8 KB
Documentation3 files · 8.8 KB
Other4 files · 1.7 MB
Repository1 file · 1.7 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 1.5 GB | c6f8580be55a |
| config.json | Configuration | 2.2 KB | — |
| union_calibration.json | Configuration | 10.7 KB | — |
| LICENSE | Documentation | 1.1 KB | — |
| README.md | Documentation | 2.2 KB | — |
| USAGE.md | Documentation | 5.5 KB | — |
| assets/figure-1.png | Other | 122.7 KB | f1937f7b6ee3 |
| assets/figure-2.png | Other | 296.9 KB | 192b3bd396e9 |
| assets/figure-3.png | Other | 254.8 KB | 055de9b7fd7f |
| freesolo-source.zip | Other | 987.5 KB | 7d3c4c6a9174 |
| .gitattributes | Repository | 1.7 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 1.5 GB
Download from Zekun Shi
Released by Zekun Shi through its official repository on Hugging Face. Read the license.
Built From
- Derived from zekun-shi/FreeSolo-L-341M
- Described by arXiv:2610.17191
- Trained on (disclosed) HuggingFaceFW/fineweb-edu
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 1.5 GB |
| 16-bit | 0.7 GB |
| 8-bit | 0.4 GB |
| 4-bit | 0.2 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About FreeSolo-B-341M
How much GPU memory does FreeSolo-B-341M need?
About 0.9 GB at 16-bit and 0.2 GB at 4-bit: the weights (374M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run FreeSolo-B-341M 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 FreeSolo-B-341M commercially?
Yes. FreeSolo-B-341M is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
Similar Models
SmolLM2Prover is a specialized, fine-tuned version of prithivMLmods/SmolLM2-CoT-360M. While retaining the strong conversational abilities of its base model, this version has been specifically enhanced to excel at deep thinking, logical reasoning, and higher-level mathematics, with a focus on generating step-by-step proofs and explanations (Chain-of-Thought). The model was fine-tuned using multiple rounds of Supervised Fine-Tuning (SFT) with the TRL library on a curated dataset, enhancing its ability to follow complex instructions and reason through problems. This model is intended to be used for text generation tasks that require logical reasoning or advanced conversation. The easiest way…
The Drafter agent of a LangGraph multi-agent support system for the fictional shop ReVonance. Base model HuggingFaceTB/SmolLM2-360M-Instruct, fine-tuned with LoRA (merged). Stage: sft (sft = supervised on the response simulator; dpo = then aligned with human and critic preferences via Direct Preference Optimisation). Knowledge-base fingerprint: f9cdb07b5df6. Demo model trained on synthetic data; it only knows the toy policies of ReVonance.
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Suggest a title and description for any text. On-device titles and descriptions: a short factual title and a one- to two-sentence description for any passage of text. Swift (requirements) Then add the Title product to your target. The MLX trait is required: without it the module compiles as a stub. Get a title and a one or two sentence description for any passage of text, on device. Fine-tuned on transcript clips, but it works on any prose. The register is deliberately plain, with no emoji, no hashtags and no clickbait, and a description is meant to identify this passage rather than its topic. An MLX model directory. Load the folder, not a single file. The chat template is not incidental. A…
The three-language base with a 2,048-token context: AnuLM-Base-400M continued for 10,000 steps at block 2,048 with YaRN, on 46M tokens of the same Hindi / English / Python proportions it was originally trained on. Five hours on one RTX 5070 Ti. Full log and the honest reading of what it bought: docs/RESULTS.md §28, with the zero-shot measurement it is compared against Hindi and English Wikipedia are CC BY-SA, C4 is ODC-BY, and the Python slice is codeparrot-clean, de-duplicated GitHub Python with mixed licences. Not affiliated with Sarvam AI, AI4Bharat, BharatGen or the Government of India. Every other checkpoint in this project is trained at 512 tokens. This one answers what happens if you…