localtune
Benchmark local LLMs on your own hardware, and find out which model, quantization and KV cache setting runs best on your machine.
Each run measures:
| Metric | Meaning |
|---|---|
| Prefill tok/s | How fast the model reads your prompt |
| Decode tok/s | How fast it writes the answer |
| TTFT ms | Time to first token: the wait before the answer starts |
| Peak VRAM GB | GPU memory used at the busiest moment |
| Avg / peak W | GPU power draw |
| Energy Wh | Electricity the GPU used during the run |
| Cost | That energy times an electricity price |
Data sources
Electricity cost uses the most local price localtune has: your zip code, then your state or country, then the US average. A price per kWh that you give yourself always replaces these. Each cost records which price it was based on.
| Price | Source | Covers | License |
|---|---|---|---|
| US zip codes | NREL, U.S. Electric Utility Companies and Rates (2023), scaled by each state's change since 2023 (EIA) | Zip codes where one company both sells and delivers power | CC BY 4.0 |
| US states | EIA-861M monthly sales and revenue by state, last 12 months | All 50 states and DC | Public domain |
| Countries | Eurostat, electricity prices for household consumers (nrg_pc_204), all taxes included, in euros | 41 European countries | CC BY 4.0 |
| US average | EIA Short-Term Energy Outlook, October 2026 | Whole US | Public domain |
The tables in src/localtune/data/ are built from these sources by
scripts/build_price_tables.py.
License
1 maintainer
Metadata
1 maintainer
Release files for localtune 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| localtune-0.1.0.tar.gz | 123.2 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| localtune-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 213.3 kB
Release files / localtune-0.1.0.tar.gz
| Download URL | localtune-0.1.0.tar.gz |
|---|---|
| Size | 123.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
b86e81a43ed1616221528ad9acdf983b000fbba252cd9ebf60d166e6b108e27b
|
|
BLAKE2b-256 checksum How to use checksums |
27043e21cafa34be0b30de63fe35e1439e94ef28991a5e596beddf50233eb445
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.12.21 {"installer":{"name":"uv","version":"0.12.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"26.04","id":"resolute","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / localtune-0.1.0-py3-none-any.whl
| Download URL | localtune-0.1.0-py3-none-any.whl |
|---|---|
| Size | 90.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
86ae5f4adb032c03f4598eb4982eeaae80eda71c36dd66dfddf267543b275e7a
|
|
BLAKE2b-256 checksum How to use checksums |
bb3e1cdc62ec285ba7210638efe97febe7a0ddda26cba9864fe67faa7bfeab5a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.12.21 {"installer":{"name":"uv","version":"0.12.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"26.04","id":"resolute","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for localtune 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
