AI chatbots now power daily work, research, and creativity, making the choice of model a practical decision with real stakes for accuracy and productivity. New releases shift the leaderboard every few months as capabilities advance. Vote to rank the strongest contenders.
What counts: Best means strongest overall intelligence and speed plus practical features at fair cost; figures are approximate.
Claude 3.5 Sonnet▲ Rising
Claude 3.5 Sonnet took the lead in mid-2024 by topping coding and reasoning benchmarks while staying fast enough for daily use. Its artful refusal of harmful requests and strong long-context performance made it the default pick for many professionals. The model earned its current spot through consistent benchmark wins rather than hype.
Intelligence: 9.4/10Speed: 8.2/10Knowledge Cutoff: Apr 2024Multimodal: ✓Monthly Price: $20
added by IJR
GPT-4o▲ Rising
OpenAI's GPT-4o unified text, vision, and voice in one fast package, instantly becoming the most widely used chatbot. Its real-time voice mode and broad plugin ecosystem gave it an edge in consumer adoption. The model holds its high rank through unmatched versatility and ecosystem maturity.
Intelligence: 9.1/10Speed: 9/10Knowledge Cutoff: Oct 2023Multimodal: ✓Monthly Price: $20
added by IJRGPT-4o
Microsoft Copilot▲ Rising
Microsoft Copilot integrated GPT-4 and DALL-E directly into Windows and Office, giving it massive reach for productivity users. Its enterprise data protection and Bing grounding helped it win business adoption. The product climbed by turning chatbot power into everyday Office workflows.
Intelligence: 8/10Speed: 8.5/10Knowledge Cutoff: 2023Multimodal: ✓Monthly Price: $30
added by IJRMicrosoft Copilot
Gemini 1.5 Pro
Google's Gemini 1.5 Pro surprised everyone with a 1-million-token context window that actually works for large codebases and books. Its native integration with Google search and YouTube gives it unique real-time grounding. The model climbed the ranks by proving long-context reliability at scale.
Intelligence: 8.9/10Speed: 7.8/10Knowledge Cutoff: Nov 2024Multimodal: ✓Monthly Price: $20
added by IJR
Mistral Large 2▲ Rising
Mistral's Large 2 model delivered European-grade performance with strong multilingual support and competitive coding scores. Its efficient architecture kept inference costs lower for developers. The chatbot gained traction by offering a capable alternative outside US labs.
Intelligence: 8.4/10Speed: 8/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $15
added by IJR
Vicuna 33B▲ Rising
Vicuna showed the power of community fine-tuning on Llama weights, delivering surprisingly capable chat at no cost. Its early open release sparked widespread experimentation. The model retains a spot for historical impact and continued community use.
Intelligence: 6.8/10Speed: 7.2/10Knowledge Cutoff: 2023Multimodal: —Monthly Price: $0
added by IJR
Perplexity Pro
Perplexity turned web search into a conversational experience with cited sources and focused research mode. Its Pro tier added GPT-4o and Claude access under one interface, winning over researchers. The service earned its place by solving the citation and freshness problems other chatbots still ignore.
Intelligence: 8.3/10Speed: 9.1/10Knowledge Cutoff: LiveMultimodal: —Monthly Price: $20
added by IJR
Gemma 2 27B▲ Rising
Google's Gemma 2 27B delivered open weights with strong safety tuning and competitive benchmark scores. Its permissive license encouraged widespread adoption by researchers and startups. The model gained rank through transparency and solid performance.
Intelligence: 7.7/10Speed: 8.6/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $0
added by IJR
Grok 2
xAI's Grok 2 brought real-time X data and a less censored personality to the chatbot space, carving out a niche for timely and humorous answers. Its rapid iteration cycle and image generation via Flux kept it competitive. The bot rose quickly by leaning into current events and user freedom.
Intelligence: 8.5/10Speed: 8.7/10Knowledge Cutoff: 2024Multimodal: ✓Monthly Price: $16
added by IJR
ChatGLM-4
Zhipu AI's ChatGLM-4 brought strong Chinese-language performance and competitive English results with competitive pricing. Its focus on long context and tool use appealed to Asian developers. The model earned its place by dominating its home market.
Intelligence: 7.8/10Speed: 7.9/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $8
added by IJR
Meta AI leveraged Llama 3.1 to deliver a free, fast chatbot inside Instagram and WhatsApp with billions of users. Its open weights strategy encouraged community fine-tunes and rapid iteration. The model rose through accessibility and zero cost rather than raw benchmark dominance.
Intelligence: 7.8/10Speed: 9.3/10Knowledge Cutoff: Dec 2023Multimodal: ✓Monthly Price: $0
added by IJRin 2 rankingsMeta AI
Qwen2 72B
Alibaba's Qwen2 72B surprised the community with top-tier math and coding results from an open model. Its strong Chinese and English performance opened new markets. The chatbot climbed by showing open models could compete on hard benchmarks.
Intelligence: 8.2/10Speed: 8.3/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $0
added by IJR
Command R+
Cohere's Command R+ focused on enterprise RAG and tool use, delivering reliable citation and long-document handling. Its optimized pricing for business workloads won contracts where generic chatbots fell short. The model rose by solving practical retrieval problems.
Intelligence: 7.9/10Speed: 7.5/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $25
added by IJR
Phi-3.5
Microsoft's Phi-3.5 showed that small, efficient models can punch above their weight on reasoning tasks when trained carefully. Its on-device capability opened new mobile and edge use cases. The model rose by proving size is not everything.
Intelligence: 7.5/10Speed: 9.4/10Knowledge Cutoff: 2024Multimodal: ✓Monthly Price: $0
added by IJR
Character.AI specialized in roleplay and personality-driven conversations, building a massive user base around creative and entertainment use. Its fine-tuned models excel at staying in character. The service earned its ranking through unmatched engagement in its niche.
Intelligence: 6.5/10Speed: 9.2/10Knowledge Cutoff: 2023Multimodal: —Monthly Price: $10
added by IJRCharacter.ai
Amazon Q brought enterprise-grade chatbot capabilities to AWS with strong security and company data grounding. Its focus on business workflows won it adoption inside large organizations. The product rose by solving internal knowledge access at scale.
Intelligence: 7.2/10Speed: 7.6/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $25
added by IJRAmazon Q
Yi-1.5 34B▲ Rising
01.AI's Yi-1.5 34B offered strong bilingual performance and competitive coding at accessible prices. Its rapid release cycle kept it relevant in the open-model race. The chatbot climbed by filling the gap between small and frontier open models.
Intelligence: 7.6/10Speed: 8.1/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $10
added by IJR
Llama 3.1 405B
Meta's open 405B model proved that open weights can match closed frontier performance on many tasks when hosted well. Platforms like Groq made it feel instant, shifting the conversation about openness. It earned its ranking by democratizing near-top-tier intelligence.
Intelligence: 8.6/10Speed: 9.5/10Knowledge Cutoff: Dec 2023Multimodal: —Monthly Price: $0
added by IJR
You.com combined multiple models with search and apps into one customizable interface, appealing to power users who want control. Its early focus on privacy and multiple chat modes differentiated it. The platform holds rank through flexibility and model choice.
Intelligence: 7.3/10Speed: 8.4/10Knowledge Cutoff: LiveMultimodal: —Monthly Price: $10
added by IJRYou.com
Pi by Inflection
Pi positioned itself as a supportive, empathetic companion with a distinct personality aimed at emotional conversations. Its focus on conversation quality over raw capability carved a unique niche. The chatbot retains ranking through consistent user loyalty in personal use.
Intelligence: 6.9/10Speed: 8.3/10Knowledge Cutoff: 2023Multimodal: —Monthly Price: $0
added by IJR
GroqChat▼ Falling
GroqChat showcased Llama and Mixtral models running at extreme speed on custom silicon, resetting expectations for response latency. Its open model hosting attracted developers testing new weights quickly. The service gained rank by making open models feel instant.
Intelligence: 7.9/10Speed: 9.8/10Knowledge Cutoff: VariesMultimodal: —Monthly Price: $0
added by IJR
HuggingChat
HuggingChat offered an open-source interface to many community models, becoming the go-to playground for trying new open weights. Its transparency and model variety supported research and experimentation. The platform holds its place through openness and accessibility.
Intelligence: 7/10Speed: 7.8/10Knowledge Cutoff: VariesMultimodal: —Monthly Price: $0
added by IJR
DeepSeek Chat▼ Falling
DeepSeek's chatbot delivered strong coding performance at very low cost, attracting developers tired of high API bills. Its rapid iteration and open releases built a loyal following. The service earned its spot through raw performance-per-dollar.
Intelligence: 8.1/10Speed: 8.8/10Knowledge Cutoff: 2024Multimodal: —Monthly Price: $5
added by IJR
Bing Chat▼ Falling
Microsoft's early Bing Chat brought GPT-4 to search with citations, proving the value of grounded responses before other chatbots caught up. Its integration with Edge gave it early visibility. The service keeps relevance through continued updates and search grounding.
Intelligence: 7.4/10Speed: 8/10Knowledge Cutoff: LiveMultimodal: —Monthly Price: $0
added by IJR
Poe by Quora▼ Falling
Poe aggregated many top models behind a single subscription, letting users compare Claude, GPT, and others instantly. Its bot creation tools attracted creators. The platform rose by solving model access fragmentation.
Intelligence: 7.5/10Speed: 8.9/10Knowledge Cutoff: VariesMultimodal: ✓Monthly Price: $20
added by IJR
Free account · your comment posts right after signup
Wild that Claude 3.5 Sonnet is winning this. Scroll down — the real answer is further down the board.
Claude 3.5 Sonnet at the top is the correct call and I'll die on this hill.
Came for the argument, stayed for the comparison grid. The numbers settle half the fights here.