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

Multiverse Computing: HyperNova 60B

4 benchmarks
CompanyMultiverse Computing
Country of originSpain
WeightsOpen weights
Parameter count60.0B
Architecturegpt-oss
VariantReasoning

Leaderboard position

#334
Overall rank (of 1121 models)
1459
Elo rating
4
Benchmarks
Top 29.8%
Percentile
#2
In Multiverse Computing (of 2 models)
Above: Prism ML: ternary-bonsai-2-27b #333 Below: imagineart: imagineart-2.0 #335

Benchmark highlights

Best benchmark
AIME 2025 76
Worst benchmark
Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning) 29.3

Nearest competitor from another company

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

appeared as nearest in 1 of 4 benchmarks

DeepSeek: deepseek-v3.2-speciale
1/4 25.0%
OpenBMB: minicpm5-2b
1/4 25.0%
Qwen: qwen3-vl-4b-thinking
1/4 25.0%
Mistral AI: Magistral Small 1.2
1/4 25.0%
Model Votes % of benchmarks
DeepSeek: deepseek-v3.2-speciale 1 25.0%
OpenBMB: minicpm5-2b 1 25.0%
Qwen: qwen3-vl-4b-thinking 1 25.0%
Mistral AI: Magistral Small 1.2 1 25.0%

Benchmarks

AIME 2025 76 MMLU-Pro 71 GPQA Diamond 60 Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning) 29.3

Predicted scores

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

Benchmark Predicted Confidence Based on
Humanity's Last Exam (no tools) ~21.74 61% OpenBMB: minicpm5-2b, Technology Innovation Institute: falcon-h1r-7b, Google: Gemma 4 12B Instruct
IFBench ~63.19 61% OpenBMB: minicpm5-2b, Technology Innovation Institute: falcon-h1r-7b, NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b
LiveCodeBench v6 ~69.41 61% Google: Gemma 4 E4B Instruct, OpenBMB: minicpm5-2b, Google: Gemma 4 12B Instruct
Artificial Analysis Intelligence Index v4.1 (Maximum Reasoning) ~25.5 61% Google: Gemma 4 E4B Instruct, Technology Innovation Institute: falcon-h1r-7b, Google: Gemma 4 12B Instruct
MATH-500 ~96.8 60% OpenBMB: minicpm5-2b, Technology Innovation Institute: falcon-h1r-7b, webai-official: twil-8b
AIME 2026 ~85.19 60% Google: Gemma 4 E4B Instruct, OpenBMB: minicpm5-2b, Google: Gemma 4 12B Instruct
HMMT Feb 2026 ~76.52 60% OpenBMB: minicpm5-2b, webai-official: twil-8b, Qwen: Qwen3 4B
IFEval ~90.32 60% OpenBMB: minicpm5-2b, webai-official: twil-8b, Qwen: Qwen3 4B
HumanEval ~91.06 60% webai-official: twil-8b, Qwen: Qwen3 4B, Tencent: youtu-llm-2b
MBPP ~93.7 60% webai-official: twil-8b, Qwen: Qwen3 4B, Tencent: youtu-llm-2b
Terminal-Bench v2.1 ~42.46 60% OpenBMB: minicpm5-2b, recursiveai: ripple-65b-preview, Xiaomi: MiMo-V2.5 Pro
SWE-bench Verified ~57.32 60% OpenBMB: minicpm5-2b, Qwen: Qwen3 4B, Arcee AI: trinity-large-thinking
SWE-bench Pro ~35.82 60% OpenBMB: minicpm5-2b, DeepSeek: deepseek-v3.2, recursiveai: ripple-65b-preview
SciCode ~36.11 60% OpenBMB: minicpm5-2b, Technology Innovation Institute: falcon-h1r-7b, NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b
AA-LCR ~54.91 60% OpenBMB: minicpm5-2b, SpaceXAI: Grok 3, SpaceXAI: Grok 3 Mini
Humanity's Last Exam (with tools) ~30.22 60% NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b, OpenAI: gpt-oss-20b, DeepSeek: deepseek-v3.2
-(CoderPPL) ~-1.69 60% Google: Gemma 4 E4B Instruct, Qwen: Qwen3 4B, Mistral AI: ministral-3-3b-instruct-2512
MMMU-Pro ~64.3 60% Google: Gemma 4 E4B Instruct, Google: Gemma 4 12B Instruct, Qwen: Qwen3.5 27B
MATH ~86.14 60% Mistral AI: ministral-3-14b-instruct-2512, Xiaomi: MiMo-V2.5 Pro
BBH ~78.42 60% Qwen: Qwen3 4B, Tencent: youtu-llm-2b, Xiaomi: MiMo-V2.5 Pro
PIQA ~84.39 60% Qwen: Qwen3.5 0.8B
BrowseComp ~55.23 60% Qwen: Qwen3.5 27B
GPQA ~66.51 60% NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b, DeepSeek: deepseek-v3.2, Moonshot AI: Kimi K2 Thinking
SuperGPQA ~54.85 60% OpenBMB: minicpm5-2b, Qwen: Qwen3.5 9B, OpenAI: gpt-oss-20b
ParseBench ~55.03 60% Google: Gemma 4 E4B Instruct, Qwen: Qwen3.5 9B, Qwen: Qwen3.5 0.8B