Olmo-3-1125-32B

Home/Models/allenai/Olmo-3-1125-32B

Olmo-3-1125-32B

allenai/Olmo-3-1125-32B

kind: original
quant: bf16 (16-bit)
family: olmo
modality: text-generation
root model: allenai/Olmo-3-1125-32B
provider: allenai
license: apache-2.0
language: en
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 40,111
likes: 124
created: —
last modified: 2025-12-03T00:26:04+00:00 (2025)
huggingface: https://huggingface.co/allenai/Olmo-3-1125-32B
tags: transformerssafetensorsolmo3text-generationendataset:allenai/dolma3_mix-5.5T-1125license:apache-2.0model-indexendpoints_compatibleregion:us

Benchmark scores (12)

benchmark	category	score	conf	evidence
PIQA	knowledge	93.3	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
WinoGrande	knowledge	87.5	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
DROP	reasoning	86.6	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
HellaSwag	knowledge	86.3	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
BBH	reasoning	81.1	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
MMLU	knowledge	79.7	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
MBPP	coding	69.8	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
MedQA	domain	68.4	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
HumanEval	coding	65.6	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
MATH	math	64.7	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
MMLU-Pro	knowledge	61.1	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`
BigCodeBench	coding	48.1	low	`\| Qwen-2.5-32B \| 64.7 \| 48.1 \| 65.6 \| 8.0 \| 43.3 \| 69.8 \| 49.7 \| 53.6 \| 48.3 \| 97.0 \| 79.7 \| 68.8 \| 68.4 \| 97.1 \| 82.2 \| 85.0 \| 88.4 \| 81.2 \| 89.9 \| 93.3 \| 86.6 \| 96.8 \| 86.6 \| 97.`

Discussions (0)

none scraped.