Lara Mini Base Report (base-model protocol) - MGSM (8 (CoT)-shot): leaderboard
Harness: lm-evaluation-harness 0.4.14.dev0; task configs unchanged except shots (A4.T17); no chat template; identical prompts/settings/answer extraction for every model (§5.1; Appendix D). Backends: Log-likelihood tasks: HF Transformers backend, fp32 log-softmax. Generative tasks: vLLM, bf16, greedy decoding, 4,096-token context; backends agree within 0.6 points on checked tasks (§5.1; Appendix D). Shots and metrics: Per A4.T17 (e.g. HellaSwag 10-shot acc_norm, MMLU 5-shot, MMLU-Pro 5-shot CoT exact match, GSM8K 5-shot EM, MATH 4-shot Math-Verify, MGSM 8-shot CoT over 11 languages, MBPP/MBPP+ 3-shot pass@1, CRUXEval 2-shot pass@1) (A4.T17). Answer extraction: Stricter parsers for GSM8K, MGSM (native CoT answer phrase) and MBPP+ (must pass MBPP and EvalPlus tests); MATH Minerva prompts with Math-Verify over 5,000 problems; CRUXEval direct prediction with official parser (Appendix D Answer extraction). Evaluated Lara checkpoint: Released checkpoint after the 100B-token long-context stage (translated/Lara-4-Mini-Base @ c998c0e1251a734eeff364d532330790a4f93c65 as captured; revision used by the authors not stated) (§5 introduction; §7 Release). Long context: RULER (13 tasks), ONERULER (4 retrieval tasks, 26 languages), HELMET (4 categories) at 32k and 128k with base-model modifications in A4.T24 (§5.7; Appendix D.3; A4.T24). Runs / seeds: not stated; scoring is log-likelihood or greedy generation (not stated). Benchmark availability: T-HellaSwag is the authors' translation, announced as "will release" at v1 (§5.2; §7). This board: Full benchmark (MATH: 5,000 problems; others not printed); Single greedy generation per item (vLLM bf16, 4,096-token context); run count not stated. lm-evaluation-harness 0.4.14.dev0, 8 (CoT)-shot (A4.T17); answer-extraction overrides in Appendix D. Metric: Exact match on each language's native CoT answer phrase, averaged over all 11 languages (%) Higher is better.
Metric: Exact match on each language's native CoT answer phrase, averaged over all 11 languages (%). Source: arxiv.org. Saturation forecast: Around February 2027. 55 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3-8B-Base | 71.9 | #1774 |
| 2 | Qwen3-4B-Base | 66.9 | #2469 |
| 3 | Hunyuan-4B-Pretrain | 63.1 | #2455 |
| 4 | Qwen3.5-4B-Base | 61.6 | #2478 |
| 5 | Qwen 2.5 7B | 60.8 | #2829 |
| 6 | Apertus-v1.5-8B-base | 54.6 | #4427 |
| 7 | Gemma 4 E4B | 52.5 | #2412 |
| 8 | Lara-4-Mini-Base | 51.7 | #5152 |
| 9 | Qwen3-1.7B-Base | 50.9 | #5246 |
| 10 | Ministral-3-3B-Base-2512 | 49.2 | #3138 |
| 11 | granite-4.1-3B-base | 47.3 | #6109 |
| 12 | Domyn-Edge-Base | 47 | #4368 |
| 13 | LFM3-1.7B-Base | 46.2 | #6594 |
| 14 | Nemotron-Flash-3B | 46.1 | #6435 |
| 15 | Falcon-H1-1.5B-Deep-Base | 44.3 | #5677 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=lara-mini-base-report-base-model-protocol-mgsm-8-cot-shot · How It Works · Data refreshed daily, snapshot 2026-11-16.