LetTokenBurnARC-Challenge — AI evaluation benchmark listed by opencompass.
$ npm i -g lettokenburn-cli$ ltb bench pull arc-challenge
External harness. pip install opencompass && opencompass --models <model_cfg> --datasets ARC_c_cot_gen_926652 (OpenCompass dataset config dir 'ARC_c') 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 | |||
|---|---|---|---|---|---|---|
| GPT-4 | 96.4 | N/A | Independent | medium | Mar 15, 2023 | arxiv.org |
| Claude 2 | 91.0 | N/A | Independent | medium | Jul 11, 2023 | www-files.anthropic.com |
| Claude 1.3 | 90.0 | N/A | Independent | medium | Jul 11, 2023 | www-files.anthropic.com |
| Claude Instant 1.1 | 85.7 | N/A | Independent | medium | Jul 11, 2023 | www-files.anthropic.com |
| GPT-3.5 Turbo | 85.2 | N/A | Independent | medium | Mar 15, 2023 | arxiv.org |
| PaLM 2-L | 69.2 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| PaLM 2-M | 64.9 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| PaLM 2-S | 59.6 | N/A | Independent | medium | May 17, 2023 | arxiv.org |
| LLaMA 33B | 57.8 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| LLaMA 65B | 56.0 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| Mistral 7B | 55.5 | N/A | Independent | medium | Oct 10, 2023 | arxiv.org |
| GPT-3-175B | 53.2 | N/A | Independent | medium | May 28, 2020 | arxiv.org |
| LLaMA 13B | 52.7 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| GPT-3 | 51.4 | N/A | Independent | medium | Nov 16, 2022 | arxiv.org |
| GPT-3-175B | 51.4 | N/A | Independent | medium | May 28, 2020 | arxiv.org |
| BLOOM 176B | 50.9 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| LLaMA 7B | 47.6 | N/A | Independent | medium | Feb 27, 2023 | arxiv.org |
| GPT-NeoX 20B | 45.4 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| phi-1.5-web 1.3B | 44.9 | N/A | Independent | medium | Sep 11, 2023 | arxiv.org |
| OPT 66B | 44.5 | N/A | Independent | medium | Mar 30, 2023 | arxiv.org |
| OPT-175B | 43.9 | N/A | Independent | medium | Jan 2, 2023 | arxiv.org |
| UL2 20B | 42.9 | N/A | Independent | medium | May 10, 2022 | arxiv.org |
| Pythia 12B | 36.8 | N/A | Independent | medium | Apr 3, 2023 | arxiv.org |
| BLOOM | 32.9 | N/A | Independent | medium | Nov 16, 2022 | arxiv.org |
| Pythia 12B | 31.8 | N/A | Independent | medium | Apr 3, 2023 | arxiv.org |
| UL2 20B | 29.8 | 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-Challenge — AI evaluation benchmark listed by opencompass.