Open-weight model
4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional
by Jack Hsieh JackHsieh/4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional
4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional is an open-weight model from Jack Hsieh, released under Apache License 2.0. It has 4B parameters. At 16-bit it needs about 9.7 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Runs On
What it takes to serve 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional (4B 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 | 8.0 GB | 9.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 4.0 GB | 4.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.0 GB | 2.4 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 23, 2026.
Model Card
Identity and Version
- Repository
- JackHsieh/4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional
- Publisher
- Jack Hsieh
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 4B parameters
- Languages
- Not stated by the source
- Revision
- 473083e8da09761a139f3bfd127842e4af25192f
- First published
- 2026-10-22
- Last updated
- 2026-10-22
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
Released by Jack Hsieh through its official repository on Hugging Face. Read the license.
Built From
- Derived from Qwen/Qwen3-4B-Base
- Trained on (disclosed) JackHsieh/4B-general-paragraph-nothink-prefill.k-8.statml-arxiv-iid.qwen3-ids.kv-tags-thought
- Trained on (disclosed) JackHsieh/statML-arxiv-iid-160M-20M
Memory Requirements
| Precision | Weights in memory |
|---|---|
| 16-bit | 8.0 GB |
| 8-bit | 4.0 GB |
| 4-bit | 2.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional
How much GPU memory does 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional need?
About 9.7 GB at 16-bit and 2.4 GB at 4-bit: the weights (4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional 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 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional commercially?
Yes. 4B-base.4B-gpnp-kv-tags-thought-thoughts.iid-40M-sans-256.stride-1.suffix-conditional 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.