Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

AMD: amd-olmo-1b

11 benchmarks
CompanyAMD
Country of originUSA

Leaderboard position

#912
Overall rank (of 1121 models)
185
Elo rating
11
Benchmarks
Top 81.4%
Percentile
#3
In AMD (of 4 models)
Above: Microsoft: gui-actor-7b-qwen2.5-vl #911 Below: Qwen: qwen2.5-coder-1.5b #913

Benchmark highlights

Best benchmark
SciQ 93.2
Worst benchmark
GSM8K 2.88

Nearest competitor from another company

The model from another company with the most similar benchmark scores.

appeared as nearest in 2 of 11 benchmarks

Apple: openelm-1_1b
2/11 18.2%
Model Votes % of benchmarks
Apple: openelm-1_1b 2 18.2%
T-Tech: t-lite-it-1.0 1 9.1%
Mistral AI: devstral-small-2505 1 9.1%
Technology Innovation Institute: falcon-h1-1.5b-instruct 1 9.1%
Apple: openelm-450m-instruct 1 9.1%
Arcee AI: trinity-nano-base 1 9.1%
Hugging Face Smol Models Research: smollm2-135m 1 9.1%
Sapbot: ToyLlama-50M 1 9.1%
Google: Gemma 3 270M Instruct 1 9.1%
Qwen: qwen2.5-0.5b 1 9.1%

Benchmarks

SciQ 93.2 PIQA 75.57 ARC-Easy 63.64 HellaSwag 63.61 WinoGrande 61.64 BoolQ 60.58 OpenBookQA 35.8 ARC-Challenge 33.7 MMLU 24.88 BBH 20.95 GSM8K 2.88

Predicted scores

Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.

Benchmark Predicted Confidence Based on
WikiText-2 (-ppl) ~-36.27 65% MWS AI: cotype-nano, Qwen: qwen2-1.5b-instruct, Google: gemma-3-1b-pt
TruthfulQA ~34.75 63% MWS AI: cotype-nano, Qwen: qwen2-1.5b-instruct, Qwen: Qwen3.5 2B
MMLU-Pro ~40.93 62% Google: gemma-3-1b-pt, Qwen: Qwen3.5 2B, Meta Llama: Llama 3.2 3B Instruct
CowQA ~43.96 62% MWS AI: cotype-nano, LiquidAI: lfm-2.5-8b-a1b, Meta Llama: Llama 3.2 3B Instruct
Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning) ~12.83 62% Qwen: Qwen3.5 2B, LiquidAI: lfm-2.5-8b-a1b, Meta Llama: Llama 3.2 3B Instruct
IFEval ~71.68 61% LiquidAI: lfm-2.5-8b-a1b, Meta Llama: Llama 3.2 3B Instruct, Hugging Face Smol Models Research: smollm2-135m-instruct
-(CoderPPL) ~-4.83 61% MWS AI: cotype-nano, Qwen: qwen2-1.5b-instruct, Qwen: Qwen3.5 2B
Artificial Analysis Intelligence Index v4.1 (Maximum Reasoning) ~5.94 61% Qwen: Qwen3.5 2B, LiquidAI: lfm-2.5-8b-a1b, Meta Llama: Llama 3.2 3B Instruct
GPQA Diamond ~42 60% LiquidAI: LFM2.5-230M, LiquidAI: lfm-2.5-350m, Qwen: Qwen3.5 0.8B
IFBench ~41.16 60% LiquidAI: lfm-2.5-8b-a1b, Meta Llama: Llama 3.2 3B Instruct, LiquidAI: LFM2.5-230M
GPQA ~27.52 60% Google: gemma-3-1b-pt, Meta Llama: Llama 3.2 3B Instruct, Qwen: Qwen3.5 0.8B
MBPP ~46.4 59% Google: gemma-3-1b-pt, Qwen: qwen-1.8b, Qwen: qwen2.5-0.5b
Artificial Analysis Intelligence Index v4.3.2 ~5 59% Qwen: Qwen3.5 2B, LiquidAI: lfm-2.5-8b-a1b, LiquidAI: LFM2.5 1.2B Instruct
MATH ~27.32 58% Google: gemma-3-1b-pt, Qwen: qwen-1.8b, Qwen: qwen2.5-0.5b
SuperGPQA ~22.7 58% Google: gemma-3-1b-pt, Qwen: Qwen3.5 0.8B, Qwen: qwen2.5-0.5b
AIME 2025 ~41.72 57% LiquidAI: lfm-2.5-8b-a1b, Microsoft: phi-4-mini-instruct, OpenBMB: minicpm5-1b
HumanEval ~52.72 57% Qwen: qwen-1.8b, Hugging Face Smol Models Research: smollm3-3b, Google: gemma-3-4b-pt
BrowseComp ~21.44 57% NON AI - НЕ ИИ: Random baseline - Случайный счёт, NON AI - НЕ ИИ: Human baseline - Человеческий результат
MATH-500 ~75.79 56% LiquidAI: lfm-2.5-8b-a1b, Microsoft: phi-4-mini-instruct, OpenBMB: minicpm5-1b
Humanity's Last Exam (no tools) ~4 56% Microsoft: phi-4-mini-instruct, NON AI - НЕ ИИ: Random baseline - Случайный счёт, NON AI - НЕ ИИ: Human baseline - Человеческий результат
ParseBench ~31.52 56% Qwen: Qwen3.5 2B, Qwen: Qwen3.5 4B, Qwen: Qwen3.5 0.8B
MMLU-Redux ~60.88 56% Google: gemma-3-1b-pt, Qwen: Qwen3.5 0.8B, Qwen: qwen2.5-0.5b
ScreenSpot-Pro ~39.89 56% Qwen: Qwen3.5 2B, Qwen: Qwen3.5 4B, Qwen: Qwen3.5 0.8B
LLMArena (llmarena.ru) (Snapshot from 19 june 2026) ~920.1 56% MWS AI: cotype-nano, T-Tech: t-lite-it-1.0
SWE-bench Verified ~21.6 55% Mistral AI: devstral-small-2505, Xiaomi: MiMo-V2.5 Pro