name string | version string | license string | canonical string | method dict | tasks list |
|---|---|---|---|---|---|
TechFuelHQ coding-eval | techfuelhq-coding-eval-v1 | https://creativecommons.org/licenses/by/4.0/ | https://techfuelhq.com/data/rtx-5080-llm-throughput/ | {
"api": "ollama /api/chat",
"options": {
"temperature": 0,
"seed": 42,
"num_ctx": 8192,
"num_predict": 4096
},
"reps": 2,
"scoring": "mechanical (executed checkers, exact rows, fullmatch lists, tool-call JSON); no judge model"
} | [
{
"id": "t1-bugfix-binary-search",
"family": "bugfix",
"prompt": "The following Python function is supposed to do what its docstring says, but it has a bug.\n\n```python\ndef find_first_ge(arr, target):\n \"\"\"Return the index of the FIRST element in sorted list arr that is\n greater than or equa... |
Local LLM Coding Evaluation on an RTX 5080
Task-level coding eval results for local models on a single RTX 5080, published with the open task set used to produce them.
Task-level results for local coding models measured on a single retail RTX 5080.
Published together with the open task set (coding-eval-tasks-v1.json) so the evaluation can be re-run or extended rather than merely cited. An eval you cannot reproduce is an anecdote.
Source and license
Canonical page, full method and change log: https://techfuelhq.com/articles/best-local-llm-coding-2026/
Licensed CC BY 4.0. Attribution: "TechFuelHQ" linking to the canonical page.
Corrections welcome. If a row is wrong it gets fixed against the primary source and the correction is logged.
- Downloads last month
- 32