LLMs Overview

Online LLMs in PapersGPT

Today, AI is advancing at a very fast pace, with new models and techniques coming out almost every week. The race shows no signs of slowing down, with the leaderboard of top AI models shifting almost weekly. This guide will help you to get an overview of the LLMs. You can use most of the LLMs in PapersGPT to chat pdf. Below is the models supported in PapersGPT:

Provider	Supported Models
OpenAI	gpt-5 \| gpt-5-min \| gpt-5-nano \| gpt-5-chat
	gpt-4.1 \| gpt-4.1-mini \| gpt-4.1-nano
	o4-mini-deep-research \| gpt-4o-mini-search-preview \| gpt-4o-search-preview
	o3 \| o4-mini \| o3-mini \| o1-mini
	gpt-4o \| chatgpt-4o-latest \| gpt-4o-mini \| gpt-4 \| gpt-4-turbo \| gpt-3.5-turbo
Gemini	2.5 Pro \| Flash Latest \| Flash Lite Latest \| 2.5 Flash \| 2.5 Flash-Lite
Gemini	2.0 Flash \| 2.0 Flash-Lite
Claude	Sonnet 4.5 \| Opus 4.1 \| Opus 4 \| Sonnet 4
	Sonnet 3.7 \| Sonnet 3.5 \| Haiku 3.5
	Opus 3 \| Sonnet 3 \| Haiku 3
DeepSeek	DeepSeek 3.2 Exp
Qwen	Qwen3-Max \| Qwen-Plus \| Qwen-Flash
	qwen3-235b-a22b-thinking-2507 \| qwen3-235b-a22b-instruct-2507
	qwen3-30b-a3b-thinking-2507 \| qwen3-30b-a3b-instruct-2507
Mistral	Magistral Medium \| Magistral small \| Mistral Medium 3 \| Mistral Small 3.2 \| Mistral Large
Kimi	kimi-k2-0905-preview \| kimi-k2-0711-preview \| kimi-k2-turbo-preview
Z.ai	glm-4.6 \| glm-4.5 \| glm-4.5-air \| glm-4.5-flash
OpenRouter	Grok 4 Fast(free) \| Claude Sonnet 4 \| Gemini 2.5 Flash \| DeepSeek V3.1(free) \| Gemini 2.0 Flash \| Gemini 2.5 Pro \| DeepSeek V3 0324 \| GPT-4.1 Mini \| GPT-5 ｜Gemini 2.5 Flash Lite \| Gemma 3 12B \| gpt-oss-20b \| Claude 3.7 Sonnet \| GPT-4o-mini \| Claude Sonnet 4.5 \| gpt-oss-120b \| DeepSeek V3.1 \| GPT-5 Mini \| GLM 4.5 \| Tongyi DeepResearch 30B A3B \| Qwen3 Next 80B A3B Thinking/Instruct \| DeepSeek R1 0528F(free) \| DeepSeek V3 0324(free) \| Qwen3 30B A3B \| Mistral Nemo \| Claude Opus 4.1 \| gpt-oss-20b(free) \| Mistral Medium 3.1 \| Kimi K2 0905 \| Qwen3 Max \| Grok 4
SiliconFlow	DeepSeek V3.1 Terminus \| DeepSeek V3.1 \| gpt-oss-120b \| gpt-oss-20b \| Kimi-K2-Instruct-0905 \| Qwen3-Next-80B-A3B-Instruct \| Ling-mini-2.0 \| Ling-flash-2.0 \| DeepSeek-V3 \| DeepSeek-R1 \| GLM-4.5 \| GLM-4.5-Air \| Qwen3-235B-A22B-Thinking-2507 \| Qwen3-235B-A22B-Instruct-2507 \| Qwen3-30B-A3B-Thinking-2507 \| Qwen3-30B-A3B-Instruct-2507 \| Kimi-K2-Instruct \| ERNIE-4.5-300B-A47B \| Hunyuan-A13B-Instruct \| MiniMax-M1-80k

Please note that OpenRouter and SiliconFlow are models hubs which supply the most popular SOTA models.

How to judge and choose a model?

With so many models available on the market, how do you choose the right one? Here are two reference platforms to consider:

* OpenRouter, which contains numerous models and is used by a large number of users. Therefore, the top 20 most-used models listed above are worth considering. However, it's best to prioritize them based on their intended use, as some models are better suited for coding but not necessarily for reading comprehension.

* lmarena.ai, a LLM benchmark platform, which judges models all by the manual comparative evaluation, and the evaluation is also a black box scoring system set up according to different usage scenarios.

In addition, if you use LLM frequently, a cost-effective economic model is also a point worth considering when choosing. For details, please see this article.