Jump to Company:
A
Alibaba
Qwen3 30B A3B Instruct 2507
Mid-sized version of Qwen3, the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
Parameters
29.9B
Context
32,768 tokens
Released
2025-08-01
API Type
nebius
This model is available for
answering and ranking.
Qwen3 30B A3B T 2507
Thinking
Tier Free
Improved Mid-sized version of Qwen3, the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
Parameters
29.9B
Context
262,144 tokens
Released
2025-08-01
API Type
nebius
This model is available for
answering and ranking.
Qwen3 Next 80B A3B Thinking
Thinking
Tier 1
Mid-sized version of Qwen3, the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
Parameters
80B
Context
32,768 tokens
Released
2025-08-01
API Type
nebius
This model is available for
answering and ranking.
Qwen3 235B A22B Instruct 2507
Thinking
Tier Free
The most powerful version of Qwen3, the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
Parameters
235B
Context
32,768 tokens
Released
2025-08-01
API Type
nebius
This model is available for
answering.
Qwen3 235B A22B T 2507
Thinking
Tier 1
Qwen3 (232Bx22B T) is a hybrid thinking + reasoning text model based on a sparse mixture-of-experts architecture. It balances efficiency, performance and quality in an inference service configured for high throughput. The high throughput configuration is the most cost efficient way to serve high volume workloads such as batch inference and distillation.
Parameters
235B
Context
256,000 tokens
Released
2025-08-01
API Type
together
This model is available for
answering and ranking.
A
Amazon
Nova 2 Lite v1.0
Powerful and cost-effective LLM for lightweight tasks. Nova 2 Lite is an advanced multimodal reasoning model that intelligently balances performance and efficiency by dynamically adjusting reasoning depth based on task complexity.
Parameters
Unknown
Context
1,000,000 tokens
Released
2025-12-02
API Type
bedrock
This model is available for
answering and ranking.
Nova Pro v1.0
Amazon Bedrock Nova Pro model. Advanced LLM for demanding tasks.
Parameters
Unknown
Context
32,768 tokens
Released
2024-05-01
API Type
bedrock
This model is available for
answering and ranking.
A
Anthropic
Claude 3.5 Haiku
Faster and cheaper version of Claude 3.5 Sonnet. Enhanced capabilities in speed, coding accuracy, and tool use. Engineered to excel in real-time applications, it delivers quick response times that are essential for dynamic tasks such as chat interactions and immediate coding suggestions.
Parameters
Unknown
Context
200,000 tokens
Released
2024-10-22
API Type
anthropic
This model is available for
answering and ranking.
Claude 3.7 Sonnet
Top of the line model by Anthropic for complex tasks. Advanced large language model with improved coding and problem-solving capabilities, particularly in front-end development and full-stack updates.
Parameters
Unknown
Context
200,000 tokens
Released
2025-02-15
API Type
anthropic
This model is available for
answering and ranking.
Claude 4 Sonnet
Thinking
Tier 1
Claude 4 Sonnet is the most cost-effective high-performance model in the Claude family with better control at coding.
Parameters
Unknown
Context
200,000 tokens
Released
2025-05-23
API Type
anthropic
This model is available for
answering and ranking.
Claude 4.1 Opus
Thinking
Tier 1
Claude 4.1 Opus is the most powerful high-performance model in the Claude family. Highest level of intelligence and capability.
Parameters
Unknown
Context
200,000 tokens
Released
2025-08-05
API Type
anthropic
This model is available for
answering.
Claude Haiku 4.5
Thinking
Tier Free
Faster and cheaper version of Claude 4.5 Sonnet. Unlike previous versions of Calude Sonnet, this model features also reasoning mode.
Parameters
Unknown
Context
200,000 tokens
Released
2025-10-16
API Type
anthropic
This model is available for
answering and ranking.
Claude 4.5 Sonnet
Thinking
Tier 1
Claude 4.5 Sonnet is the most cost-effective high-performance model in the Claude family. It demonstrates advancements in agent capabilities, with enhanced performance in tool handling, memory management, context processing, code generation and analysis, from identifying optimal improvements to exercising stronger judgment in refactoring decisions.
Parameters
Unknown
Context
1,000,000 tokens
Released
2025-09-29
API Type
anthropic
This model is available for
answering and ranking.
Claude 4.5 Opus
Thinking
Tier 1
Claude 4.5 Opus is the most powerful high-performance model in the Claude family. It is a powerful model for complex tasks and reasoning.
Parameters
Unknown
Context
1,000,000 tokens
Released
2025-11-01
API Type
anthropic
This model is available for
answering.
D
DeepSeek
DeepSeek V3 0324
The most recent version of DeepSeek's advanced language model V3. It demonstrates notable improvements over its predecessor, DeepSeek-V3, in several key aspects.
Parameters
671B
Context
128,000 tokens
Released
2024-06-01
API Type
nebius
This model is available for
answering and ranking.
DeepSeek R1 0528
Thinking
Tier 1
R1 0528 is the latest version of the celebrated DeepSeek R1 model. It delivers top performance on math, code, and logic tasks. It is especially good at tasks like code review, document analysis, planning, information extraction, and coding.
Parameters
671B
Context
163,840 tokens
Released
2025-05-28
API Type
nebius
This model is available for
answering.
G
Gemma 3 27B
Google's 27B parameter open source model, fast and performing. This model handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling.
Parameters
27B
Context
120,000 tokens
Released
2024-03-12
API Type
nebius
This model is available for
answering and ranking.
Gemma 3N E4B
Gemma 3n models use selective parameter activation technology to reduce resource requirements. This technique allows the models to operate at an effective size of 2B and 4B parameters, which is lower than the total number of parameters they contain.
Parameters
8B
Context
32,000 tokens
Released
2025-07-03
API Type
together
This model is available for
answering and ranking.
Gemini 2.5 Flash Lite
Thinking
Tier Free
Gemini 2.5 Flash Lite is a high-performance model that is even faster and cheaper than Gemini 2.5 Flash.
Parameters
Unknown
Context
1,048,576 tokens
Released
2025-07-01
API Type
gemini
This model is available for
answering and ranking.
Gemini 2.5 Flash
Thinking
Tier 1
Gemini 2.5 Flash is a high-performance model that is optimized for speed and cost-effectiveness. It is a good choice for tasks that require a balance of performance and cost.
Parameters
Unknown
Context
1,048,576 tokens
Released
2025-07-01
API Type
gemini
This model is available for
answering and ranking.
Gemini 2.5 Pro
Thinking
Tier 1
Highl capable model from the Gemini family. Improved quality, especially for world knowledge, code, and long context
Parameters
Unknown
Context
2,097,152 tokens
Released
2025-07-01
API Type
gemini
This model is available for
answering and ranking.
Gemini 3 Flash Preview
Thinking
Tier 1
Gemini 3 Flash combines Gemini 3 Pro's reasoning capabilities with the Flash line's levels on latency, efficiency, and cost. It not only enables everyday tasks with improved reasoning, but is designed to tackle the most complex agentic workflows.
Parameters
Unknown
Context
1,048,576 tokens
Released
2025-12-17
API Type
gemini
This model is available for
answering and ranking.
Gemini 3 Pro Preview
Thinking
Tier 1
The most advanced reasoning Gemini model, capable of solving complex problems. Gemini 3 Pro can comprehend vast datasets and challenging problems from different information sources, including text, audio, images, video, PDFs, and even entire code repositories with its 1M token context window.
Parameters
Unknown
Context
1,048,576 tokens
Released
2025-11-18
API Type
gemini
This model is available for
answering and ranking.
M
Meta
Llama 3.1 8B
Fast and cheap version of Llama 3.1 70B. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture.
Parameters
8B
Context
128,000 tokens
Released
2024-07-24
API Type
nebius
This model is available for
answering and ranking.
Llama 3.3 70B
Llama 3.3 multilingual LLM is a pretrained and instruction tuned generative model in 70B. The Llama 3.3 instruction tuned text only model is optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.
Parameters
70B
Context
128,000 tokens
Released
2024-04-18
API Type
nebius
This model is available for
answering and ranking.
Llama 4 Scout 17B (17Bx16E)
Fastest version of the Llama 4 collection of models. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding.
Parameters
109B
Context
10,000,000 tokens
Released
2024-4-5
API Type
together
This model is available for
answering and ranking.
Llama 4 Maverick 17B (17Bx128E)
The most powerful version of the Llama 4 collection of models. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding.
Parameters
400B
Context
1,000,000 tokens
Released
2024-4-5
API Type
together
This model is available for
answering and ranking.
M
MiniMax
MiniMax M2.1
Thinking
Tier Free
MiniMax M2.1 is a powerful and cost-effective LLM for advanced tasks.MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world capability while maintaining exceptional latency, scalability, and cost efficiency.
Parameters
229B
Context
196,608 tokens
Released
2025-12-23
API Type
nebius
This model is available for
answering.
M
Mistral AI
Mixtral 8x7B
Mixtral 8x7B Instruct v0.1 is a Mixture of Experts model with just 8x7B parameters. It is a good model for quick tasks and low cost.
Parameters
56B
Context
128,000 tokens
Released
2024-5-22
API Type
together
This model is available for
answering and ranking.
Mistral Small 24B
Mistral Small ( 2501 ) claims to be a high performance model in the small Large Models category below 70B, boasting 24B parameters and achieving state-of-the-art capabilities comparable to larger models.
Parameters
24B
Context
65,536 tokens
Released
2025-01-11
API Type
together
This model is available for
answering and ranking.
M
MoonShot AI
Kimi K2 Instruct
Updated version of the Kimi K2 Instruct model. It is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters. Trained with the Muon optimizer, Kimi K2 achieves exceptional performance across frontier knowledge, reasoning, and coding tasks while being meticulously optimized for agentic capabilities
Parameters
1T
Context
256,000 tokens
Released
2025-09-05
API Type
nebius
This model is available for
answering and ranking.
Kimi K2 Thinking
Reasoning version of the Kimi K2 model. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift.
Parameters
1T
Context
256,000 tokens
Released
2025-11-06
API Type
nebius
This model is available for
answering.
N
NVIDIA
Llama 3.1 Nemotron Ultra 253B
Thinking
Tier 1
Nemotron Ultra 253B is an LLM which is a derivative of Llama 3.1 70B. It is a reasoning model that is post trained for reasoning, human chat preferences, and tasks, such as RAG and tool calling.
Parameters
253B
Context
131,072 tokens
Released
2025-4-7
API Type
nebius
This model is available for
answering and ranking.
Nemotron Nano V2 12B
Thinking
Tier Free
A 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence. It introduces a hybrid Transformer-Mamba architecture, combining transformer-level accuracy with Mamba’s memory-efficient sequence modeling for significantly higher throughput and lower latency.
Parameters
12B
Context
131,072 tokens
Released
2025-10-28
API Type
nebius
This model is available for
answering and ranking.
Nemotron 3 Nano 30B A3B
Thinking
Tier Free
A small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully open with open-weights, datasets and recipes so developers can easily customize, optimize, and deploy the model on their infrastructure for maximum privacy and security.
Parameters
30B
Context
131,072 tokens
Released
2025-12-14
API Type
nebius
This model is available for
answering and ranking.
O
OpenAI
GPT-5
Thinking
Tier 1
The new top of the line OpenAi reasoning model, excelling at complex instructions and software engineering. With a 400k token context window, it is optimized for precise code diffs, agent reliability, and high recall in large documents, suiting it for agents, IDE tools, and enterprise knowledge retrieval.
Parameters
Unknown
Context
400,000 tokens
Released
2025-08-07
API Type
openai
This model is available for
answering and ranking.
GPT-5.2
Thinking
Tier 1
The new OpenAi flagship model for coding and agentic tasks across industries, excelling at complex instructions and software engineering. With a 400k token context window, it is optimized for precise code diffs, agent reliability, and high recall in large documents, suiting it for agents, IDE tools, and enterprise knowledge retrieval.
Parameters
Unknown
Context
400,000 tokens
Released
2025-12-11
API Type
openai
This model is available for
answering and ranking.
GPT-5 Mini
Thinking
Tier 1
Faster and cost-effective version of GPT-5, the new top of the line OpenAi reasoning model. It's great for well-defined tasks and precise prompts.
Parameters
Unknown
Context
400,000 tokens
Released
2025-08-07
API Type
openai
This model is available for
answering and ranking.
GPT-5 Nano
Thinking
Tier Free
The fastest and cheapest version of GPT-5, the new top of the line OpenAi reasoning model. It's great for summarization and classification tasks.
Parameters
Unknown
Context
400,000 tokens
Released
2025-08-07
API Type
openai
This model is available for
answering and ranking.
o1
Thinking
Tier 1
The full reasoning model from OpenAi's o1 family of reasoning models, the o1-series. Still powerful (and expensive), it has since been succeeded by the more advanced o3 and o4 series.
Parameters
Unknown
Context
200,000 tokens
Released
2024-12-17
API Type
openai
This model is available for
answering.
o3
Thinking
Tier 1
The top reasoning model from OpenAi's family of reasoning models. Extremely powerful (and expensive) in all tasks.
Parameters
Unknown
Context
128,000 tokens
Released
2025-01-22
API Type
openai
This model is available for
answering.
o4 Mini
Thinking
Tier 1
A fast, cost-efficient reasoning model that balances performance with affordability. O4-mini shows strong performance in coding, math, and visual tasks, making it ideal for high-volume applications needing good-enough logic and faster response times without the higher cost of larger models.
Parameters
Unknown
Context
200,000 tokens
Released
2025-04-16
API Type
openai
This model is available for
answering and ranking.
gpt oss 120b
Thinking
Tier Free
Gpt oss 120b is the first open weight model by OpenAI, a 120-billion-parameter open-weight language model for general-purpose tasks, deployable on custom infrastructure.
Parameters
120B
Context
128,000 tokens
Released
2025-08-05
API Type
nebius
This model is available for
answering and ranking.
gpt oss 20b
Thinking
Tier Free
Gpt oss 20b is the light version of the open weight model by OpenAI, a 20-billion-parameter open-weight language model for fast and cost-effective general-purpose tasks, deployable on custom infrastructure.
Parameters
20B
Context
128,000 tokens
Released
2025-08-05
API Type
nebius
This model is available for
answering and ranking.
Z
Zai Org
GLM 4.5 Air
Thinking
Tier Free
The GLM-4.5 series models are foundation models designed for intelligent agents. GLM-4.5 has 355 billion total parameters with 32 billion active parameters, while GLM-4.5-Air adopts a more compact design with 106 billion total parameters and 12 billion active parameters. GLM-4.5 models unify reasoning, coding, and intelligent agent capabilities to meet the complex demands of intelligent agent applications.
Parameters
106B
Context
128,000 tokens
Released
2025-07-20
API Type
nebius
This model is available for
answering and ranking.
GLM 4.7
Thinking
Tier 1
GLM-4.7 is Z.AI's latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while delivering more natural conversational experiences and superior front-end aesthetics.
Parameters
358.8B
Context
202,800 tokens
Released
2025-12-22
API Type
together
This model is available for
answering and ranking.
x
xAI
Grok-4 1 Fast Reasoning
Thinking
Tier Free
The most affordable reasoning model from the Grok family. It is a state of the art reasoning model trained with reinforcement learning. It delivers strong performance on math, code, and logic tasks. It is especially good at tasks like code review, document analysis, planning, information extraction, and coding.
Parameters
Unknown
Context
2,000,000 tokens
Released
2025-06-10
API Type
grok
This model is available for
answering and ranking.
Grok-4 1 Fast Non-Reasoning
The most affordable non-reasoning model from the Grok family. It is a state of the art non-reasoning model trained with reinforcement learning. It delivers strong performance on general tasks.
Parameters
Unknown
Context
2,000,000 tokens
Released
2025-06-10
API Type
grok
This model is available for
answering and ranking.
Grok-4
Thinking
Tier 1
Grok-4 is the most performing model in the Grok family. Ranking top in almost all bencmarks, it delivers strong performance on math, code, and logic tasks. It is especially good at tasks like code review, document analysis, planning, information extraction, and coding.
Parameters
Unknown
Context
256,000 tokens
Released
2025-06-10
API Type
grok
This model is available for
answering and ranking.
Grok-3
Thinking
Tier 1
Grok-3 is a powerful reasoning model from the Grok family. It is a state of the art reasoning model trained with reinforcement learning. It delivers strong performance on math, code, and logic tasks. It is especially good at tasks like code review, document analysis, planning, information extraction, and coding.
Parameters
Unknown
Context
131,072 tokens
Released
2025-03-27
API Type
grok
This model is available for
answering and ranking.
Grok-3 Mini
Thinking
Tier Free
Grok-3 mini is the more affordable version of Grok-3 by xAI.
Parameters
Unknown
Context
131,072 tokens
Released
2025-03-27
API Type
grok
This model is available for
answering and ranking.