Datasets:

jablonkagroup
/

ChemBench-Results

model_id stringlengths 5 59	name stringlengths 5 45	number_params int64 17 405B ⌀	description stringlengths 4 584	date_published stringlengths 10 10 ⌀	paper_link stringlengths 12 94 ⌀	code_link stringlengths 12 85 ⌀	is_api_endpoint bool 2 classes	nr_of_tokens int64 128k 15,000B ⌀	architecture stringclasses 5 values	is_open_weights bool 2 classes	is_open_dataset bool 2 classes	is_mixture_of_experts bool 2 classes	model_alignment stringclasses 5 values	reinforcement_learning_from_human_feedback bool 2 classes	domain_specific_pretraining bool 2 classes	domain_specific_finetuning bool 1 class	tool_use bool 2 classes	tool_type stringclasses 4 values	temperature float64 0 1 ⌀	epochs int64 0 4 ⌀	reasoning_model bool 2 classes	reasoning_type stringclasses 3 values	overall_score float64 0.02 0.86	Analytical Chemistry float64 0 0.91	Chemical Preference float64 0 0.84	General Chemistry float64 0.02 0.97	Inorganic Chemistry float64 0 0.96	Materials Science float64 0 0.91	Organic Chemistry float64 0.01 0.94	Physical Chemistry float64 0.02 0.9	Technical Chemistry float64 0 0.85	Toxicity and Safety float64 0.01 0.77	verified bool 2 classes	uses_harness bool 2 classes	harness_description stringclasses 3 values	precision stringclasses 1 value	weight_type stringclasses 1 value	base_model stringclasses 1 value
NexusSciAgent-base-1767146262	NexusSciAgent	null	ClosedSource Models (Claude-4.5-Sonnet)	null	not provided	not provided	false	null	not provided	true	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.860455	0.914474	0.842	0.966443	0.956512	0.908883	0.94411	0.896407	0.849996	0.771852	false	false		not provided	not provided	not provided
NexusSciAgent_DeepSeek-base-1767172310	NexusSciAgent_DeepSeek	null	DeepSeek_v3	null	not provided	https://github.com/CASIA-LM/NexusSciAgent	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.711603	0.578947	0.677	0.872483	0.847817	0.795178	0.873777	0.799783	0.724996	0.602963	false	false		not provided	not provided	not provided
deepseek-chat1-base-1762475946	deepseek-chat1	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.656743	0.625	0.59041	0.919463	0.869565	0.77381	0.853488	0.848485	0.825	0.478519	false	false		not provided	not provided	not provided
Haiyue-Ch1-base-1756454865	Haiyue-Ch1	null	not provided	null	not provided	not provided	false	null	Transformer decoder-only	false	false	false	not provided	false	false	false	false	not provided	0	0	true	not provided	0.648316	0.677632	0.56	0.885906	0.891304	0.75	0.827506	0.818182	0.8	0.485926	false	false		not provided	not provided	not provided
groq_meta-llama/llama-4-maverick-17b-128e-instruct_e8a077d8	meta-llama/llama-4-maverick-17b-128e-instruct	17	Llama 4 Maverick is Meta's natively multimodal model that enables text and image understanding. With a 17 billion parameter mixture-of-experts architecture (128 experts), this model offers industry-leading performance for multimodal tasks like natural assistant-like chat, image recognition, and coding tasks. With a 128...	2025-05-05	https://www.llama.com/llama-downloads/	https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct	true	128,000	e.g., transformer, mamba	false	false	false		false	false	false	false		null	null	false		0.64537	0.532895	0.594	0.892617	0.826087	0.678571	0.832168	0.739394	0.75	0.515556	true	false		not provided	not provided	not provided
o1-preview	o1-preview	null	o1 is trained with reinforcement learning and chain-of-thought reasoning to improve safety, robustness, and reasoning capabilities. The family includes o1-preview and o1-mini versions.	2024-09-12	https://cdn.openai.com/o1-system-card-20240917.pdf	null	true	null	DecoderOnly	false	false	null	null	true	false	false	false	null	1	null	true	medium	0.643472	0.625	0.563437	0.932886	0.902174	0.72619	0.830233	0.848485	0.85	0.475556	true	false		not provided	not provided	not provided
gemini-2.5-flash-preview-04-17-base-1750491557	gemini-2.5-flash-preview-04-17	null	gemini-2.5-flash-preview-04-17	null	not provided	not provided	true	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.641136	0.618421	0.593	0.852349	0.836957	0.75	0.825175	0.745455	0.775	0.475556	true	false		not provided	not provided	not provided
GPT-5-nano-base-1764459351	GPT-5-nano	null	not provided	null	not provided	not provided	true	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.634339	0.743421	0.569	0.812081	0.858696	0.738095	0.818182	0.830303	0.725	0.451852	true	false		not provided	not provided	not provided
gpt-oss-120b-base-1760293341	gpt-oss-120b	null	OpenAI's open-weight reasoning models	null	https://openai.com/index/introducing-gpt-oss/	https://huggingface.co/openai/gpt-oss-120b	true	null	not provided	true	false	false	RLVR	false	false	false	false	not provided	0	0	false	not provided	0.631371	0.618421	0.595	0.85906	0.891304	0.738095	0.818182	0.781818	0.75	0.426667	true	false		not provided	not provided	not provided
claude3.5	Claude-3.5 (Sonnet)	null	Claude models are general purpose large language models. They use a transformer architecture and are trained via unsupervised learning, RLHF, and Constitutional AI (including both a supervised and Reinforcement Learning (RL) phase). Claude 3.5 was developed by Anthropic.	2024-06-20	null	null	true	null	null	false	false	null	null	true	false	false	false	null	0	null	false	null	0.625538	0.565789	0.584416	0.825503	0.836957	0.714286	0.825581	0.769697	0.85	0.44	true	false		not provided	not provided	not provided
DeepSeek-V3-0324-W8A8-1-base-1764234320	DeepSeek-V3-0324-W8A8-1	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.625269	0.546053	0.598	0.838926	0.847826	0.714286	0.811189	0.842424	0.825	0.411852	false	false		not provided	not provided	not provided
claude3.5-react	Claude-3.5 (Sonnet) React	null	Claude models are general purpose large language models. They use a transformer architecture and are trained via unsupervised learning, RLHF, and Constitutional AI (including both a supervised and Reinforcement Learning (RL) phase). Claude 3.5 was developed by Anthropic.	2024-06-20	null	null	true	null	null	false	false	null	null	true	false	false	true	ArXiV, Web search, Wikipedia, Wolfram alpha calculator, SMILES to IUPAC name and IUPAC name to SMILES converters	0	null	false	null	0.624821	0.578947	0.599401	0.872483	0.804348	0.678571	0.837209	0.757576	0.8	0.408889	true	true	ReAct-style harness: interleaves chain-of-thought reasoning with tool-use actions, observing each tool's result before deciding the next step, rather than answering in a single forward pass. Yao et al. 2022, https://arxiv.org/abs/2210.03629	not provided	not provided	not provided
deepseek-v3-1-w4a8-1-base-1764234501	deepseek-v3-1-w4a8-1	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.621321	0.605263	0.571	0.825503	0.826087	0.690476	0.818182	0.812121	0.775	0.437037	false	false		not provided	not provided	not provided
DeepSeek-V3-0324-MTP1-base-1762579014	DeepSeek-V3-0324-MTP1	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.617732	0.526316	0.574	0.85906	0.869565	0.72619	0.825175	0.806061	0.8	0.413333	false	false		not provided	not provided	not provided
gpt-oss-20b-base-1760359009	gpt-oss-20b	null	OpenAI's open-weight reasoning models	null	https://openai.com/index/introducing-gpt-oss/	not provided	false	null	transformer	true	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.61163	0.519737	0.588	0.899329	0.869565	0.72619	0.797203	0.775758	0.725	0.38963	true	false		not provided	not provided	not provided
gpt-4o	GPT-4o	null	GPT-4o is OpenAI's third major iteration of their popular large multimodal model, GPT-4, which expands on the capabilities of GPT-4 with Vision.	2024-05-13	null	null	true	null	DecoderOnly	false	false	null	null	true	false	false	false	null	0	null	false	null	0.610832	0.559211	0.589411	0.805369	0.804348	0.75	0.755814	0.715152	0.75	0.441481	true	false		not provided	not provided	not provided
kimi-k2-instruct-0905-base-1760364223	kimi-k2-instruct-0905	null	Kimi K2 0905 is Moonshot AI's improved version of the Kimi K2 model, featuring enhanced coding capabilities with superior frontend development and tool calling performance.	null	https://arxiv.org/abs/2507.20534	https://github.com/MoonshotAI/Kimi-K2	false	null	transformer	true	false	false	RLVR	false	false	false	false	not provided	0	0	false	not provided	0.596554	0.578947	0.518	0.832215	0.771739	0.75	0.81352	0.763636	0.775	0.432593	true	false		not provided	not provided	not provided
llama3.1-405b-instruct	Llama-3.1-405B-Instruct	405,000,000,000	The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform...	2024-07-23	https://arxiv.org/abs/2407.21783	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	0	null	false	null	0.579268	0.506579	0.54046	0.791946	0.771739	0.654762	0.755814	0.709091	0.7	0.419259	true	false		not provided	not provided	not provided
mistral-large-2-123b	Mistral-Large-2	123,000,000,000	Mistral Large 2 has a 128k context window and supports dozens of languages and along with 80+ coding languages. Mistral Large 2 is designed for single-node inference with long-context applications in mind.	2024-07-24	null	https://huggingface.co/mistralai/Mistral-Large-Instruct-2407	true	null	DecoderOnly	true	false	false	null	null	false	false	false	null	0	null	false	null	0.569943	0.480263	0.546454	0.785235	0.793478	0.666667	0.732558	0.690909	0.675	0.395556	true	false		not provided	not provided	not provided
claude3	Claude-3 (Opus)	null	Claude models are general purpose large language models. They use a transformer architecture and are trained via unsupervised learning, RLHF, and Constitutional AI (including both a supervised and Reinforcement Learning (RL) phase). Claude 3 was developed by Anthropic.	2024-03-04	https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf	null	true	null	null	false	false	null	null	true	false	false	false	null	0	null	false	null	0.569584	0.467105	0.565435	0.765101	0.793478	0.630952	0.695349	0.648485	0.7	0.41037	true	false		not provided	not provided	not provided
paper-qa	PaperQA2	null	PaperQA2 is a package for doing high-accuracy retrieval augmented generation (RAG) on PDFs or text files, with a focus on the scientific literature. We used PaperQA2 via the non-public API deployed by FutureHouse and the default settings (using Claude-3.5-Sonnet as summarizing and answer-generating LLM).	2024-09-11	https://storage.googleapis.com/fh-public/paperqa/Language_Agents_Science.pdf	https://github.com/Future-House/paper-qa	false	null	null	null	null	null	null	null	null	null	true	Paper Search, Gather Evidence, Generate Answer, Citation Traversal	0	null	false	null	0.568867	0.460526	0.563437	0.724832	0.73913	0.690476	0.67907	0.678788	0.7	0.423704	true	true	PaperQA agent: retrieval-augmented QA over full-text scientific literature. It retrieves candidate papers/passages, assesses their relevance, gathers evidence, and synthesizes a cited answer — iterating retrieval and evidence steps rather than answering in one shot. Lála et al. 2023, https://arxiv.org/abs/2312.07559	not provided	not provided	not provided
llama3.1-70b-instruct	Llama-3.1-70B-Instruct	70,000,000,000	The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform...	2024-07-23	https://arxiv.org/abs/2407.21783	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	0	null	false	null	0.533716	0.407895	0.519481	0.691275	0.771739	0.666667	0.662791	0.642424	0.65	0.383704	true	false		not provided	not provided	not provided
Qwen-2.5-32B-base-1740920072	Qwen-2.5-32B	null	Qwen-2.5-32B is Alibaba's flagship model, delivering near-instant responses with GPT-4 level capabilities across a wide range of tasks.	null	https://arxiv.org/abs/2502.13923	https://github.com/QwenLM/Qwen2.5-VL	true	null	not provided	true	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.528715	0.427632	0.529	0.671141	0.684783	0.619048	0.638695	0.654545	0.575	0.383704	true	false		not provided	not provided	not provided
llama3-70b-instruct	Llama-3-70B-Instruct	70,000,000,000	Llama 3 models were trained on a text corpus of over 15T tokens. These models use a tokenizer with a vocabulary of 128K tokens. Additionally, improvements in the post-training procedures substantially reduced false refusal rates, improved alignment, and increased diversity in model responses.	2024-04-18	null	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	0	null	false	null	0.517934	0.414474	0.532468	0.604027	0.663043	0.630952	0.632558	0.593939	0.625	0.368889	true	false		not provided	not provided	not provided
llama3-70b-instruct-T-one	Llama-3-70B-Instruct (Temperature 1.0)	70,000,000,000	Llama 3 models were trained on a text corpus of over 15T tokens. These models use a tokenizer with a vocabulary of 128K tokens. Additionally, improvements in the post-training procedures substantially reduced false refusal rates, improved alignment, and increased diversity in model responses.	2024-04-18	null	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	1	null	false	null	0.516499	0.375	0.53047	0.604027	0.684783	0.619048	0.632558	0.6	0.6	0.373333	true	false		not provided	not provided	not provided
llama3.1-70b-instruct-T-one	Llama-3.1-70B-Instruct (Temperature 1.0)	70,000,000,000	The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform...	2024-07-23	https://arxiv.org/abs/2407.21783	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	1	null	false	null	0.511119	0.368421	0.535465	0.66443	0.695652	0.654762	0.553488	0.557576	0.55	0.38963	true	false		not provided	not provided	not provided
gpt-4o-react	GPT-4o React	null	GPT-4o is OpenAI's third major iteration of their popular large multimodal model, GPT-4, which expands on the capabilities of GPT-4 with Vision.	2024-05-13	null	null	true	null	DecoderOnly	false	false	null	null	true	false	false	true	ArXiV, Web search, Wikipedia, Wolfram alpha calculator, SMILES to IUPAC name and IUPAC name to SMILES converters	0	null	false	null	0.50538	0.467105	0.420579	0.758389	0.728261	0.559524	0.718605	0.6	0.725	0.374815	true	true	ReAct-style harness: interleaves chain-of-thought reasoning with tool-use actions, observing each tool's result before deciding the next step, rather than answering in a single forward pass. Yao et al. 2022, https://arxiv.org/abs/2210.03629	not provided	not provided	not provided
GPT-4o-mini-base-1764283510	GPT-4o-mini	null	GPT-4o-mini is a lightweight, cost-efficient variant of OpenAI’s GPT-4o model. It prioritizes speed and low-cost inference while retaining general-purpose language and reasoning capabilities. This benchmark was run via the official OpenAI API (text completion mode), without any fine-tuning or tool augmentation.	null	not provided	https://colab.research.google.com/drive/1n-4SU824oBlLApJRLTuM3Z_zyReP0RQX?usp=sharing	true	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.50323	0.375	0.51	0.597315	0.684783	0.488095	0.629371	0.581818	0.625	0.371852	true	false		not provided	not provided	not provided
gemma-2-9b-it	Gemma-2-9B-it	9,000,000,000	Gemma is a family of lightweight, open models built from the research and technology that Google used to create the Gemini models. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.	2024-06-27	https://arxiv.org/abs/2408.00118	https://github.com/google-deepmind/gemma	true	8,000,000,000,000	DecoderOnly	true	false	false	PPO	true	false	false	false	null	0	null	false	null	0.482425	0.315789	0.551449	0.543624	0.554348	0.52381	0.555814	0.484848	0.525	0.339259	true	false		not provided	not provided	not provided
gemma-2-9b-it-T-one	Gemma-2-9B-it (Temperature 1.0)	9,000,000,000	Gemma is a family of lightweight, open models built from the research and technology that Google used to create the Gemini models. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.	2024-06-27	https://arxiv.org/abs/2408.00118	https://github.com/google-deepmind/gemma	true	8,000,000,000,000	DecoderOnly	true	false	false	PPO	true	false	false	false	null	1	null	false	null	0.480273	0.289474	0.557443	0.557047	0.543478	0.5	0.546512	0.466667	0.475	0.342222	true	false		not provided	not provided	not provided
phi-3-medium-4k-instruct	Phi-3-Medium-4k-Instruct	14,000,000,000	The Phi-3-Medium-4K-Instruct is a 14B parameters, lightweight, state-of-the-art open model trained with the Phi-3 datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. The model belongs to the Phi-3 family with the Mediu...	2024-05-21	https://arxiv.org/abs/2404.14219	https://huggingface.co/microsoft/Phi-3-medium-4k-instruct	false	4,800,000,000,000	DecoderOnly	true	false	false	null	true	false	false	false	null	0	null	false	null	0.474534	0.342105	0.532468	0.47651	0.630435	0.547619	0.560465	0.460606	0.55	0.331852	true	false		not provided	not provided	not provided
claude2	Claude-2	null	Claude models are general purpose large language models. They use a transformer architecture and are trained via unsupervised learning, RLHF, and Constitutional AI (including both a supervised and Reinforcement Learning (RL) phase). Claude 2 was developed by Anthropic.	2023-07-11	https://www-cdn.anthropic.com/bd2a28d2535bfb0494cc8e2a3bf135d2e7523226/Model-Card-Claude-2.pdf	null	true	null	null	false	false	null	null	true	false	false	false	null	0	null	false	null	0.473458	0.375	0.511489	0.503356	0.608696	0.464286	0.593023	0.50303	0.475	0.331852	true	false		not provided	not provided	not provided
llama3.1-8b-instruct	Llama-3.1-8B-Instruct	8,000,000,000	The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform...	2024-07-23	https://arxiv.org/abs/2407.21783	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	0	null	false	null	0.471664	0.394737	0.527473	0.503356	0.5	0.404762	0.581395	0.509091	0.45	0.325926	true	false		not provided	not provided	not provided
gpt-3.5-turbo	GPT-3.5 Turbo	null	GPT-3.5 Turbo, developed by OpenAI, features a context window of 4096 tokens.	2023-11-06	null	null	true	null	DecoderOnly	false	false	null	null	true	false	false	false	null	0	null	false	null	0.466284	0.381579	0.534466	0.489933	0.543478	0.47619	0.588372	0.4	0.4	0.30963	true	false		not provided	not provided	not provided
ChemPile-LoRA-Ensemble-Qwen2.5-7B-Instruct-1753950951	ChemPile-LoRA-Ensemble	null	not provided	null	not provided	not provided	false	null	not provided	true	true	false	not provided	false	true	false	false	not provided	0	0	false	not provided	0.46626	0.355263	0.517	0.47651	0.630435	0.452381	0.559441	0.436364	0.575	0.334815	true	false		not provided	not provided	not provided
llama3.1-8b-instruct-T-one	Llama-3.1-8B-Instruct (Temperature 1.0)	8,000,000,000	The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform...	2024-07-23	https://arxiv.org/abs/2407.21783	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	1	null	false	null	0.461263	0.361842	0.523477	0.530201	0.478261	0.416667	0.576744	0.418182	0.4	0.32	true	false		not provided	not provided	not provided
llama3-8b-instruct-T-one	Llama-3-8B-Instruct (Temperature 1.0)	8,000,000,000	Llama 3 models were trained on a text corpus of over 15T tokens. These models use a tokenizer with a vocabulary of 128K tokens. Additionally, improvements in the post-training procedures substantially reduced false refusal rates, improved alignment, and increased diversity in model responses.	2024-04-18	null	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	1	null	false	null	0.4566	0.401316	0.52048	0.436242	0.543478	0.452381	0.551163	0.345455	0.625	0.324444	true	false		not provided	not provided	not provided
llama3-8b-instruct	Llama-3-8B-Instruct	8,000,000,000	Llama 3 models were trained on a text corpus of over 15T tokens. These models use a tokenizer with a vocabulary of 128K tokens. Additionally, improvements in the post-training procedures substantially reduced false refusal rates, improved alignment, and increased diversity in model responses.	2024-04-18	null	https://github.com/meta-llama/llama3	true	15,000,000,000,000	DecoderOnly	true	false	false	DPO	true	false	false	false	null	0	null	false	null	0.455524	0.407895	0.515485	0.442953	0.48913	0.416667	0.562791	0.369697	0.6	0.324444	true	false		not provided	not provided	not provided
gemini-pro	Gemini-Pro	null	Gemini models are built from the ground up for multimodality: reasoning seamlessly across text, images, audio, video, and code.	2024-06-07	https://arxiv.org/abs/2403.05530	null	true	null	DecoderOnly	false	false	null	null	true	false	false	false	null	0	null	false	null	0.452654	0.388158	0.5005	0.483221	0.467391	0.5	0.567442	0.448485	0.475	0.308148	true	false		not provided	not provided	not provided
command-r+	Command-R+	104,000,000,000	Cohere Command R is a family of highly scalable language models that balance high performance with strong accuracy. Command-R models were released by Cohere.	2024-04-04	null	https://huggingface.co/CohereForAI/c4ai-command-r-plus	true	null	null	false	false	null	null	true	false	false	false	null	0	null	false	null	0.447633	0.342105	0.513487	0.496644	0.521739	0.464286	0.551163	0.327273	0.5	0.311111	true	false		not provided	not provided	not provided
qwen2.5-instruct-base-1751944608	qwen2.5-instruct	null	not provided	null	not provided	not provided	false	null	not provided	true	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.427136	0.243421	0.504	0.436242	0.565217	0.392857	0.503497	0.369697	0.45	0.302222	true	false		not provided	not provided	not provided
inspect_file_test-base-1768655468	inspect_file_test	null	inspect_file_test	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.4257	0.407895	0.498	0.409396	0.456522	0.428571	0.526807	0.327273	0.625	0.26963	true	false		not provided	not provided	not provided
inspect_test_2-base-1768660541	inspect_test_2	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.4257	0.407895	0.498	0.409396	0.456522	0.428571	0.526807	0.327273	0.625	0.26963	true	false		not provided	not provided	not provided
mixtral-8x7b-instruct	Mixtral-8x7b-Instruct	47,000,000,000	Mixtral is a sparse mixture-of-experts network. It is a decoder-only model where the feedforward block picks from a set of 8 distinct groups of parameters. At every layer, for every token, a router network chooses two of these groups (the “experts”) to process the token and combine their output additively.	2023-12-11	https://arxiv.org/abs/2401.04088	https://huggingface.co/mistralai/Mixtral-8x7B-v0.1	true	null	DecoderOnly	true	false	true	DPO	false	false	false	false	null	0	null	false	null	0.42396	0.269737	0.535465	0.422819	0.554348	0.416667	0.47907	0.333333	0.325	0.26963	true	false		not provided	not provided	not provided
kunlun-70b-1-base-1764233793	kunlun-70b-1	null	not provided	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.422111	0.223684	0.534	0.38255	0.565217	0.428571	0.41958	0.509091	0.4	0.271111	true	false		not provided	not provided	not provided
mixtral-8x7b-instruct-T-one	Mixtral-8x7b-Instruct (Temperature 1.0)	47,000,000,000	Mixtral is a sparse mixture-of-experts network. It is a decoder-only model where the feedforward block picks from a set of 8 distinct groups of parameters. At every layer, for every token, a router network chooses two of these groups (the “experts”) to process the token and combine their output additively.	2023-12-11	https://arxiv.org/abs/2401.04088	https://huggingface.co/mistralai/Mixtral-8x7B-v0.1	true	null	DecoderOnly	true	false	true	DPO	false	false	false	false	null	1	null	false	null	0.418221	0.276316	0.522478	0.449664	0.51087	0.404762	0.472093	0.345455	0.325	0.266667	true	false		not provided	not provided	not provided
gpt-4	GPT-4	null	GPT-4 is a large multimodal model released by OpenAI to succeed GPT-3.5 Turbo. It features a context window of 32k tokens.	2023-03-14	https://arxiv.org/abs/2303.08774	null	true	null	DecoderOnly	false	false	null	null	null	false	false	false	null	0	null	false	null	0.412841	0.427632	0.163836	0.697987	0.695652	0.607143	0.67907	0.642424	0.7	0.41037	true	false		not provided	not provided	not provided
llama2-70b-chat	Llama-2-70B Chat	70,000,000,000	LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. Llama was released by Meta AI.	2023-07-18	https://arxiv.org/abs/2302.13971	https://huggingface.co/meta-llama/Llama-2-70b-chat-hf	false	2,000,000,000,000	DecoderOnly	true	false	false	null	true	false	false	false	null	0.01	null	false	null	0.266141	0.072368	0.487512	0.134228	0.217391	0.178571	0.146512	0.169697	0.125	0.136296	true	false		not provided	not provided	not provided
llama2-13b-chat	Llama-2-13B Chat	13,000,000,000	LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. Llama was released by Meta AI.	2023-07-18	https://arxiv.org/abs/2302.13971	https://huggingface.co/meta-llama/Llama-2-13b-chat-hf	false	2,000,000,000,000	DecoderOnly	true	false	false	null	true	false	false	false	null	0	null	false	null	0.25538	0.092105	0.484515	0.114094	0.271739	0.095238	0.153488	0.151515	0.1	0.100741	true	false		not provided	not provided	not provided
olympiad-base-1741075285	olympiad	null	test	null	not provided	not provided	false	null	not provided	false	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.219547	0.157895	0.371	0.107383	0.217391	0.190476	0.181818	0.078788	0.25	0.093333	true	false		not provided	not provided	not provided
gemma-1-1-7b-it	Gemma-1.1-7B-it	7,000,000,000	Gemma is a family of lightweight, open models built from the research and technology that Google used to create the Gemini models. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.	2024-02-21	https://arxiv.org/abs/2403.08295	https://github.com/google-deepmind/gemma	true	6,000,000,000,000	DecoderOnly	true	false	false	PPO	true	false	false	false	null	0	null	false	null	0.192253	0.210526	0.004995	0.33557	0.413043	0.357143	0.37907	0.290909	0.375	0.22963	true	false		not provided	not provided	not provided
gemma-1-1-7b-it-T-one	Gemma-1.1-7B-it (Temperature 1.0)	7,000,000,000	Gemma is a family of lightweight, open models built from the research and technology that Google used to create the Gemini models. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning.	2024-02-21	https://arxiv.org/abs/2403.08295	https://github.com/google-deepmind/gemma	true	6,000,000,000,000	DecoderOnly	true	false	false	PPO	true	false	false	false	null	1	null	false	null	0.190818	0.210526	0.008991	0.348993	0.413043	0.357143	0.372093	0.30303	0.375	0.216296	true	false		not provided	not provided	not provided
Llama-2-7b-chat-hf -base-1741273893	Llama-2-7b-chat-hf	null	LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. Llama was released by Meta AI.	null	https://arxiv.org/abs/2307.09288	https://huggingface.co/meta-llama/Llama-2-7b-chat-hf	false	null	not provided	true	false	false	not provided	false	false	false	false	not provided	0	0	false	not provided	0.040208	0.006579	0.003	0.020134	0	0	0.020979	0.024242	0.025	0.014815	true	false		not provided	not provided	not provided
galactica_120b	Galactica-120b	120,000,000,000	Galactica is a large language model developed by Facebook. It is a transformer-based model trained on a large corpus of scientific data.	2022-11-01	https://galactica.org/paper.pdf	https://huggingface.co/facebook/galactica-120b	false	450,000,000,000	DecoderOnly	true	false	false	null	false	true	false	false	null	0	4	false	null	0.015067	0	0	0.046053	0.053191	0	0.011338	0.055866	0	0.023188	true	false		not provided	not provided	not provided

README.md exists but content is empty.

Downloads last month: 219

Number of rows:

189

Total file size:

97.8 kB

Space using jablonkagroup/ChemBench-Results 1

Collection including jablonkagroup/ChemBench-Results

ChemBench-Collection

Collection

Datasets, Spaces and Results related to ChemBench • 4 items • Updated Oct 3, 2025 • 4