AIBase
EN

AI News

View More

World's First! China's TongGeometry Model Achieves Autonomous Problem Creation and Second-Level Solving in Geometric Olympiad

11.5k 49 minutes ago
World's First! China's TongGeometry Model Achieves Autonomous Problem Creation and Second-Level Solving in Geometric Olympiad

GPT5.2 Shocked the Mathematical Community: Solving 11 Erdos Problems in Half a Month, Tao TeXiang Comments

13.9k 1 hours ago
GPT5.2 Shocked the Mathematical Community: Solving 11 Erdos Problems in Half a Month, Tao TeXiang Comments

AI Power Showdown: OpenAI and Google Both Win Gold in International Mathematical Olympiad!

5.1k 21 hours ago
AI Power Showdown: OpenAI and Google Both Win Gold in International Mathematical Olympiad!

Microsoft Launches Phi-4 Generative AI Model with Improved Mathematical Abilities and Training Data Quality

11.1k 01-15
Microsoft Launches Phi-4 Generative AI Model with Improved Mathematical Abilities and Training Data Quality

AI Products

View More

DeepScaleR-1.5B-Preview

A large language model optimized by reinforcement learning, focusing on enhancing mathematical problem-solving skills.

Learning and education
13.3k

Eurus-2-7B-SFT

Eurus-2-7B-SFT is a large language model optimized for mathematical capabilities, focusing on reasoning and problem-solving.

Research tools
9.4k

EurusPRM-Stage2

EurusPRM-Stage2 is a reinforcement learning model based on implicit process rewards aimed at enhancing the reasoning capabilities of generative models.

Model training and deployment
7.7k

EurusPRM-Stage1

EurusPRM-Stage1 is a reinforcement learning model based on implicit process rewards, aimed at enhancing the reasoning abilities of generative models.

AI model
9.2k

Models

View More

Qianfan-VL-70B

Baidu

Qianfan-VL-70B

-

Input tokens/M

-

Output tokens/M

32

Context Length

Qianfan-VL-8B

Baidu

Qianfan-VL-8B

-

Input tokens/M

-

Output tokens/M

32

Context Length

Hunyuan-T1-latest

Tencent

Hunyuan-T1-latest

$1

Input tokens/M

$4

Output tokens/M

32

Context Length

Hunyuan-Large-Vision

Tencent

Hunyuan-Large-Vision

-

Input tokens/M

-

Output tokens/M

24

Context Length

Doubao-Seed-1.6-thinking

Bytedance

Doubao-Seed-1.6-thinking

$0.8

Input tokens/M

$8

Output tokens/M

256

Context Length

Spark X1

Iflytek

Spark X1

$2

Input tokens/M

-

Output tokens/M

-

Context Length

Doubao-1.5-thinking-pro

Bytedance

Doubao-1.5-thinking-pro

$4

Input tokens/M

$16

Output tokens/M

128

Context Length

DeepSeek-V3

Deepseek

DeepSeek-V3

$2

Input tokens/M

$8

Output tokens/M

32

Context Length

Gemini 2.5 Pro

Google

Gemini 2.5 Pro

$8.75

Input tokens/M

$70

Output tokens/M

1k

Context Length

QianfanHuijin-8B

Baidu

QianfanHuijin-8B

-

Input tokens/M

-

Output tokens/M

32

Context Length

Qianfan-QI-VL

Baidu

Qianfan-QI-VL

-

Input tokens/M

-

Output tokens/M

32

Context Length

Claude Opus 4

Anthropic

Claude Opus 4

$105

Input tokens/M

$525

Output tokens/M

200

Context Length

Gemini Diffusion

Google

Gemini Diffusion

-

Input tokens/M

-

Output tokens/M

-

Context Length

Qianfan-Llama-VL-8B

Baidu

Qianfan-Llama-VL-8B

-

Input tokens/M

-

Output tokens/M

32

Context Length

o3

Openai

o3

$14

Input tokens/M

$56

Output tokens/M

200

Context Length

Gemma 3 4B

Google

Gemma 3 4B

$0.14

Input tokens/M

$0.28

Output tokens/M

131

Context Length

GPT-4.5

Openai

GPT-4.5

$525

Input tokens/M

$1050

Output tokens/M

128

Context Length

Grok-3 Mini

Xai

Grok-3 Mini

$2.1

Input tokens/M

$3.5

Output tokens/M

128

Context Length

Grok-3

Xai

Grok-3

$21

Input tokens/M

$105

Output tokens/M

128

Context Length

DeepSeek-R1-Distill-Qwen-32B

Deepseek

DeepSeek-R1-Distill-Qwen-32B

$2

Input tokens/M

-

Output tokens/M

8

Context Length

Qwen 1.5B_THIP

Thrillcrazyer

Q

Qwen-1.5B_THIP is a mathematical reasoning model fine-tuned using the GRPO method on the DeepMath-103k mathematical dataset based on DeepSeek-R1-Distill-Qwen-1.5B with the TRL framework. This model is specifically optimized for solving mathematical problems and has strong mathematical reasoning abilities.

Natural Language ProcessingTransformersTransformers
Thrillcrazyer
492
2

Palmyra Mini Thinking AIO GGUF

prithivMLmods

P

The palmyra-mini series of models demonstrates excellent capabilities in the fields of complex reasoning and mathematical problem-solving, especially performing outstandingly in benchmark tests that require in-depth understanding and multi-step thinking. This model is proficient in solving elementary school-level math problems, achieving a high score of 0.818 in the gsm8k benchmark test, also obtaining a score of 0.818 in the MATH500 benchmark test, and a stable score of 0.6 in the AMC23 benchmark test.

Natural Language ProcessingTransformersTransformersMultiple Languages
prithivMLmods
2.1k
2

MobileLLM R1 950M Base

facebook

M

MobileLLM-R1 is a series of efficient inference language models released by Meta, focusing on solving mathematical, programming, and scientific problems. This model can still achieve excellent performance with a relatively small parameter scale, and provides a complete training recipe and data sources to support reproducible research.

Natural Language ProcessingTransformersTransformersEnglish
facebook
302
10

MobileLLM R1 140M Base

facebook

M

MobileLLM-R1 is a series of efficient inference models launched by Facebook, focusing on solving mathematical, programming, and scientific problems. This model has achieved excellent performance in multiple benchmark tests with only about 2T high-quality labels for pre-training.

Natural Language ProcessingTransformersTransformersEnglish
facebook
336
8

MobileLLM R1 360M

facebook

M

MobileLLM-R1 is a series of efficient reasoning models launched by Meta, focusing on solving mathematical, programming, and scientific problems. The model offers three scale versions of 140M, 360M, and 950M, with excellent reasoning ability and reproducibility.

Natural Language ProcessingTransformersTransformersEnglish
facebook
328
10

AceMath 1.5B Instruct

nvidia

A

Cutting-edge mathematical reasoning model in the AceMath series, excelling at solving English math problems using Chain-of-Thought (CoT) reasoning.

Natural Language ProcessingSafetensorsSafetensorsEnglish
nvidia
718
9

Castula U2 QwenRe 1.5B

prithivMLmods

C

A compact multilingual reasoning model fine-tuned based on Qwen-1.5B, excelling in mathematical problem solving, logical reasoning, code generation, and general tasks

Natural Language ProcessingTransformersTransformersMultiple Languages
prithivMLmods
35
2

OpenVLThinker 7B

ydeng9

O

OpenVLThinker-7B is a vision-language reasoning model specifically designed for multimodal tasks, with particular optimization for solving visual mathematical problems.

MultimodalTransformersTransformers
ydeng9
594
16

Gemma 2 2b It Grpo Gsm8k

lmassaron

G

A mathematical reasoning model fine-tuned on the GSM8k dataset using the GRPO method based on Google/gemma-2-2b-it, specifically optimized for solving mathematical problems

Natural Language ProcessingTransformersTransformers
lmassaron
948
2

AceMath 7B Instruct

nvidia

A

AceMath-7B-Instruct is a specialized instruction model for mathematical reasoning developed by NVIDIA, based on an improved Qwen architecture, excelling at solving English math problems through chain-of-thought (CoT) reasoning.

Natural Language ProcessingSafetensorsSafetensorsEnglish
nvidia
1.5k
22

SocraticLM

CogBase-USTC

S

An education-oriented large language model fine-tuned based on Qwen2.5-Math-7B-Instruct, supporting Socratic guidance and mathematical problem solving

Natural Language ProcessingTransformersTransformersMultiple Languages
CogBase-USTC
118
9

Rho Math 1b V0.1

microsoft

R

Rho-1 is a language model specialized in mathematics, pretrained with Selective Language Modeling (SLM) method, significantly improving accuracy in solving mathematical problems.

Natural Language ProcessingTransformersTransformersEnglish
microsoft
1.5k
15

LeerooDedicated Math 7b

leeroo

L

This model is constructed through expert collaboration methods, specializing in mathematical problem solving, capable of autonomously generating solutions or invoking GPT-4 level large models when necessary.

Natural Language ProcessingTransformersTransformers
leeroo
63
6

OpenMath CodeLlama 7b Python Hf

nvidia

O

The OpenMath model is specifically designed for solving mathematical problems by integrating textual reasoning with Python interpreter-executed code blocks. Trained on the OpenMathInstruct-1 dataset containing 1.8 million math problem-solution pairs.

Natural Language ProcessingTransformersTransformersMultiple Languages
nvidia
83
7

Deepseek Math 7b Rl

deepseek-ai

D

DeepSeekMath is a large language model specialized in solving mathematical problems, capable of handling complex mathematical reasoning and computation tasks.

Natural Language ProcessingTransformersTransformers
deepseek-ai
790
82

Deepseek Math 7b Base

deepseek-ai

D

DeepSeekMath is a large language model specialized in solving mathematical problems, featuring powerful mathematical reasoning and computational capabilities.

Natural Language ProcessingTransformersTransformers
deepseek-ai
6.9k
71

Parallel_7B

Mathoctopus

P

MathOctopus is a multilingual mathematical reasoning large language model based on the LLaMA 2 architecture, supporting 10 languages and specializing in solving mathematical problems.

Natural Language ProcessingTransformersTransformersMultiple Languages
Mathoctopus
14
2

Tora Code 13b V1.0

llm-agents

T

ToRA-7B is a 7B-parameter tool-integrated reasoning agent, specifically designed for solving mathematical problems by combining natural language reasoning with external tool usage.

Natural Language ProcessingTransformersTransformersMultiple Languages
llm-agents
128
15

MathCoder CL 7B

MathLLMs

M

The MathCoder series of open-source large language models, specifically designed for general mathematical problem solving, fine-tuned based on Llama-2

Natural Language ProcessingTransformersTransformersEnglish
MathLLMs
74
18

MathCoder L 7B

MathLLMs

M

The MathCoder series of open-source large language models, specifically tailored for general mathematical problem solving, fine-tuned based on Llama-2 and Code Llama.

Natural Language ProcessingTransformersTransformersEnglish
MathLLMs
127
18
AIBase
Empowering the future, your artificial intelligence solution think tank
English简体中文繁體中文にほんご
FirendLinks:
AI Newsletters AI ToolsMCP ServersAI NewsAIBaseLLM LeaderboardAI Ranking
© 2026AIBase
Business CooperationSite Map