LetTokenBurnARC-Easy — AI evaluation benchmark listed by opencompass.
$ npm i -g lettokenburn-cli$ ltb bench pull arc-easy
External harness. pip install opencompass && opencompass --models <model_cfg> --datasets ARC_e_gen (OpenCompass dataset config dir 'ARC_e') https://github.com/open-compass/opencompass
License is link-only: ltb bench pull fetches rows from the original host for local evaluation only and does not write a solution/ directory. Do not redistribute the generated tasks.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
| Harness | Source | Confidence | Link | |||
|---|---|---|---|---|---|---|
| PaLM 2-L | 89.7 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| PaLM 2-M | 88.0 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| PaLM 2-S | 85.6 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| Mixtral 8x7B | 83.1 | N/A | Independent | medium | Jan 8, 2024 | arxiv.org |
| Mistral 7B | 80.5 | N/A | Independent | medium | Jan 8, 2024 | arxiv.org |
| Mistral 7B | 80.0 | N/A | Independent | medium | Oct 10, 2023 | arxiv.org |
| LLaMA 33B | 80.0 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| LLaMA 65B | 78.9 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| phi-1.5-web 1.3B | 76.1 | N/A | Independent | medium | Sep 11, 2023 | arxiv.org |
| BLOOM 176B | 75.9 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| LLaMA 13B | 74.8 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| LLaMA 7B | 72.8 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| Pythia 12B | 71.5 | N/A | Independent | medium | Apr 3, 2023 | arxiv.org |
| OPT 66B | 71.3 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| GPT-3-175B | 71.2 | N/A | Independent | medium | May 28, 2020 | arxiv.org |
| OPT-175B | 71.0 | N/A | Independent | medium | Jan 2, 2023 | arxiv.org |
| GPT-NeoX 20B | 70.8 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| Pythia 12B | 70.2 | N/A | Independent | medium | Apr 3, 2023 | arxiv.org |
| GPT-3 | 68.8 | N/A | Independent | medium | Nov 16, 2022 | arxiv.org |
| GPT-3-175B | 68.8 | N/A | Independent | medium | May 28, 2020 | arxiv.org |
| BLOOM | 40.7 | N/A | Independent | medium | Nov 16, 2022 | arxiv.org |
| UL2 20B | 38.4 | N/A | Independent | medium | May 10, 2022 | arxiv.org |
| UL2 20B | 32.2 | N/A | Independent | medium | May 10, 2022 | arxiv.org |
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
ARC-Easy — AI evaluation benchmark listed by opencompass.