LetTokenBurn
Sign in
Incomplete29%Harbor: pendingKnowledge & Reasoning

Balrog

Tracked by Epoch AI's Benchmarking Hub (results collected from external sources).

Leaderboard

How to run

ltb · balrog
$ npm i -g lettokenburn-cli
$ ltb bench info balrog

External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.

Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs

Scores

39 sourced entries
HarnessSourceConfidenceLink
GPT-6 Astra68.3N/AIndependentmediumSep 18, 2026balrogai.com
Claude Opus 563.4N/AIndependentmediumSep 20, 2026balrogai.com
GPT-5.6 Sol60.0N/AIndependentmediumSep 18, 2026balrogai.com
Gemini 3 Pro58.1N/AIndependentmediumFeb 3, 2026balrogai.com
Gemini 3.1 Pro57.0N/AIndependentmediumFeb 25, 2026balrogai.com
GPT-5.6-terra53.2N/AIndependentmediumSep 20, 2026balrogai.com
Gemini 3 Flash Preview48.1N/AIndependentmediumFeb 13, 2026balrogai.com
GPT-5.6 Luna45.6N/AIndependentmediumSep 18, 2026balrogai.com
Grok 4 070943.6N/AIndependentmediumJul 23, 2025balrogai.com
Claude Opus 4.5 (2025-11-01)43.5N/AIndependentmediumFeb 24, 2026balrogai.com
Gemini 2.5 Pro Exp 03-2543.3N/AIndependentmediumApr 25, 2025balrogai.com
claude-opus-4-5-20251101-64K43.0N/AIndependentmediumFeb 24, 2026balrogai.com
DeepSeek R134.9N/AIndependentmediumApr 10, 2025balrogai.com
Gemini 2.5 Flash33.5N/AIndependentmediumJul 22, 2025balrogai.com
GPT-532.8N/AIndependentmediumAug 19, 2025balrogai.com
Claude 3.5 Sonnet32.6N/AIndependentmediumNov 11, 2024balrogai.com
GPT-4o32.3N/AIndependentmediumNov 11, 2024balrogai.com
Claude Haiku 4.531.2N/AIndependentmediumFeb 24, 2026balrogai.com
claude-haiku-4-5-20251001-1K31.2N/AIndependentmediumFeb 24, 2026balrogai.com
Grok 329.5N/AIndependentmediumApr 25, 2025balrogai.com
Reka Flash 329.2N/AIndependentmediumMar 13, 2025balrogai.com
Llama-3.1-70B-Instruct27.9N/AIndependentmediumNov 11, 2024balrogai.com
Llama-3.2-90B-Vision-Instruct27.3N/AIndependentmediumNov 11, 2024balrogai.com
Llama-3.3-70B-Instruct23.0N/AIndependentmediumDec 9, 2024balrogai.com
gemini-1.5-pro-00221.0N/AIndependentmediumNov 11, 2024balrogai.com
DeepSeek-R1-Distill-Qwen-32B19.5N/AIndependentmediumJan 26, 2025balrogai.com
Claude 3.5 Haiku (2024-10-22)19.3N/AIndependentmediumDec 11, 2024balrogai.com
Mistral-Nemo-Instruct-240717.6N/AIndependentmediumDec 9, 2024balrogai.com
GPT-4o-mini17.4N/AIndependentmediumNov 11, 2024balrogai.com
Llama-3.2-11B-Vision-Instruct16.8N/AIndependentmediumNov 11, 2024balrogai.com
Qwen2.5-72B-Instruct16.2N/AIndependentmediumNov 25, 2024balrogai.com
Llama-3.1-8B-Instruct15.1N/AIndependentmediumNov 11, 2024balrogai.com
gemini-1.5-flash-00214.6N/AIndependentmediumNov 11, 2024balrogai.com
Qwen2-VL-72B12.8N/AIndependentmediumNov 25, 2024balrogai.com
Phi-411.6N/AIndependentmediumJan 13, 2025balrogai.com
Llama-3.2 3B10.1N/AIndependentmediumNov 11, 2024balrogai.com
Qwen2.5-7B-Instruct7.8N/AIndependentmediumNov 25, 2024balrogai.com
Llama-3.2 1B6.6N/AIndependentmediumNov 11, 2024balrogai.com
Qwen2-VL-7B3.7N/AIndependentmediumNov 25, 2024balrogai.com

Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.

Reproduced by the community

All runs
Loading community runs…

About

Tracked by Epoch AI's Benchmarking Hub (results collected from external sources). Random baseline 0; score ceiling 1 (0–1 scale).