functiongemma-langcalc-en.zh is an open-weight model from Jeffrey Dick, released under Gemma Terms of Use. It has 268M parameters and a 32,768-token context. At 16-bit it needs about 0.6 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 112 downloads a month.
A fine-tuned model based on google/functiongemma-270m-it.
Runs On
What it takes to serve functiongemma-langcalc-en.zh (268M 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 | 0.5 GB | 0.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.3 GB | 0.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 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 18, 2026.
functiongemma-langcalc-en.zh on every accelerator the SAVRN Index prices, at every precision
Model Card
A fine-tuned model based on google/functiongemma-270m-it.
Excerpt from the card by Jeffrey Dick, licensed gemma.
Configuration
- Architecture
- Gemma3ForCausalLM
- Context length (tokens)
- 32,768
- Layers
- 18
- Hidden size
- 640
- Feed-forward size
- 2,048
- Attention heads
- 4
- Key/value heads
- 1
- Head dimension
- 256
- Vocabulary size
- 262,144
- Sliding window (tokens)
- 512
- Model type
- gemma3_text
Identity and Version
- Repository
- jedick/functiongemma-langcalc-en.zh
- Publisher
- Jeffrey Dick
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 268M parameters
- Languages
- en, zh
- Revision
- 0e340fd33d06aa6d4219ad166263f7ea02d04f14
- First published
- 2026-09-20
- Last updated
- 2026-10-11
Files and Weights
9 files, 854.8 MB in total. The weights are 1 file totalling 536.2 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 536.2 MB | 708fc42d50de |
| config.json | Configuration | 1.5 KB | — |
| generation_config.json | Configuration | 225 B | — |
| README.md | Documentation | 423 B | — |
| chat_template.jinja | Other | 13.8 KB | — |
| langcalc_q8_ekv1024.litertlm | Other | 285.2 MB | 272bd1554db2 |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 33.4 MB | 80d7f800b949 |
| tokenizer_config.json | Tokenizer | 821 B | — |
License and Download
- License
- gemma
- Access
- Open weights, no gate
- Download size
- 536.2 MB
Released by Jeffrey Dick through its official repository on Hugging Face.
Built From
- Derived from google/functiongemma-270m-it
- Trained on (disclosed) jedick/langcalc-en
- Trained on (disclosed) jedick/langcalc-zh
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 536.2 MB |
| 16-bit | 0.5 GB |
| 8-bit | 0.3 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About functiongemma-langcalc-en.zh
How much GPU memory does functiongemma-langcalc-en.zh need?
About 0.6 GB at 16-bit and 0.2 GB at 4-bit: the weights (268M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run functiongemma-langcalc-en.zh 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 functiongemma-langcalc-en.zh commercially?
Yes, with conditions. functiongemma-langcalc-en.zh is released under Gemma Terms of Use. Gemma models are released under Google's Gemma Terms of Use, which permit commercial use and redistribution subject to the Gemma Prohibited Use Policy, whose restrictions must be passed on to anyone the model is distributed to.
What is functiongemma-langcalc-en.zh's context length?
32,768 tokens, from the maximum position embeddings in its published configuration.