Claude Opus 5.5 for Nepali sentiment analysis on five-point scale
Claude Opus 5.5 outperforms encoders on a five-point Nepali sentiment analysis task.
From arXiv cs.AI
Model picks
Which model for which job right now. Expires fast. 137 on file.
Coding71Analysis and data57Research55Agents and automation49Writing47Learning and teaching12Planning and PM12Personal and everyday8ChatGPT59Claude43Gemini14Copilot9Codex6Grok1
More model picks
Just added
- Claude Opus 5.5 for Nepali sentiment analysis on five-point scaleClaude Opus 5.5 outperforms encoders on a five-point Nepali sentiment analysis task. · arXiv cs.AI
- LLMs for zero-shot Nepali sentiment analysis on three-class benchmarkCommercial LLMs show strong zero-shot performance on Nepali sentiment analysis. · arXiv cs.AI
- Claude Opus for cross-component protocol vulnerability exploitationClaude Opus 4.8 shows high success rates in discovering and exploiting protocol vulnerabilities in complex systems. · arXiv cs.AI
- Claude Haiku 5.5 for coding and finance tasksUse Claude Haiku 5.5 for coding agents, financial analysis, and long work. · TLDR AI
All 137
137 entries
LLMs for zero-shot Nepali sentiment analysis on three-class benchmark
Commercial LLMs show strong zero-shot performance on Nepali sentiment analysis.
From arXiv cs.AI
Claude Opus for cross-component protocol vulnerability exploitation
Claude Opus 4.8 shows high success rates in discovering and exploiting protocol vulnerabilities in complex systems.
From arXiv cs.AI
GPT-6 Pro for difficult tasks and long workflows
GPT-6 Pro, powered by GPT-6 Astra, is the top option for challenging tasks and extended processes.
From OpenAI Help Center
Use GPT-5.6 Luna for free users
Free and Go users will use GPT-5.6 Luna as the default model for everyday questions.
From OpenAI Help Center
Model selection for Pollo AI's video creation workflow
Pollo AI uses a mix of GPT-5.6 and GPT-6 Astra to balance reasoning depth, speed, and cost for different video creation tasks.
From OpenAI News
Claude Haiku 5.5 for coding and finance tasks
Use Claude Haiku 5.5 for coding agents, financial analysis, and long work.
From TLDR AI
Claude Haiku 5.5 for short prompts under 100k tokens
Use Claude Haiku 5.5 for tasks under 100,000 tokens where it matches GPT-6 Luna pricing and performance.
From Simon Willison
Use Claude Haiku 5.5 for fast coding tasks in GitHub Copilot
Claude Haiku 5.5 is now available in GitHub Copilot for fast, high-volume coding tasks.
From GitHub Changelog: Copilot
Claude Haiku 5.5 for knowledge work and terminal tasks
Claude Haiku 5.5 excels at knowledge work and terminal tasks, offering strong performance with a large context window.
From Artificial Analysis
Claude Sonnet 4 leads in geospatial AI agent evaluation
Claude Sonnet 4 shows the highest capability in geospatial and environmental tasks evaluated by GeoNatureAgent.
From arXiv cs.AI
Claude Sonnet 4.6 and InternVL3.5-8B for Spatial Reasoning Analysis
Claude Sonnet 4.6 and InternVL3.5-8B show distinct failure modes in spatial reasoning benchmarks.
From arXiv cs.AI
Claude Sonnet 4 for sentence-level text simplification
Claude Sonnet 4 is a strong choice for sentence-level text simplification tasks, achieving high scores in evaluations.
From arXiv cs.AI
GPT-6 model choice for reasoning and task type
Match GPT-6 model and reasoning effort to workload for balance of capability, cost, and latency.
From OpenAI News
Claude Opus 4.5 for Text-to-MQL tasks
Claude Opus 4.5 shows strong performance on the AptMQL-Bench for text-to-MQL translation.
From arXiv cs.AI
Gemini 4 Argon for complex reasoning and cybersecurity
Gemini 4 Argon excels at deep reasoning, complex tasks, and cybersecurity defense, offering competitive performance at a lower cost.
From Engadget AI
Gemini 4 Argon for benchmark performance and long context
Gemini 4 Argon leads in benchmarks and offers a 1M token output, suitable for complex tasks.
From Latent Space
GPT-6.1 Sol for coding and professional work
Use GPT-6.1 Sol for complex coding and professional tasks when cost is a factor compared to Astra.
From Model Documentation
Gemini 3.1 Pro Preview for Enhanced Reasoning and Token Efficiency
Leverage Gemini 3.1 Pro Preview for improved reasoning, token efficiency, and factually consistent experiences in software engineering and agent workflows.
From Gemini Models
Gemini 3.8 Flash-Lite TTS for low latency text to speech
Use Gemini 3.8 Flash-Lite TTS for everyday tasks needing fast, natural-sounding speech synthesis.
From Gemini Models
Use GPT-6.1 Sol for Agentic Coding and Terminal Workflows
GPT-6.1 Sol offers strong multistep coding and efficient token use for agentic and terminal tasks.
From GitHub Changelog: Copilot
GPT-6.1 Sol for professional tasks at lower cost
Use GPT-6.1 Sol for coding, document analysis, and business workflows when cost is a factor but performance is key.
From OpenAI News
Claude Sonnet 5.5 for claude.ai free tier
Use Claude Sonnet 5.5 for the free tier on claude.ai for a more capable free offering.
From Simon Willison
Choose Gemini 3.8 Flash-Lite TTS for production workloads
Select Gemini 3.8 Flash-Lite TTS for high-throughput production needs, replacing older preview models.
From Gemini Models
Claude Sonnet-4.6 for traffic control intent translation
Claude Sonnet-4.6 achieves high accuracy in translating traffic shaping intents to Linux tc configurations.
From arXiv cs.AI
Use Claude Opus 5.5 in GitHub Copilot for coding tasks
Claude Opus 5.5 is now available in GitHub Copilot for agentic coding and knowledge work.
From GitHub Changelog: Copilot
GPT-6 Astra for demanding Context Graph queries
Use GPT-6 Astra for the most challenging graph-query tasks, especially with messy, real-world data.
From OpenAI News
Claude Opus 5 for exploit development
Use Claude Opus 5 for developing exploits due to its advanced capabilities.
From TechCrunch AI
Gemini 3.8 Live Extended Thinking for complex agent tasks
Use Gemini 3.8 Live Extended Thinking for enterprise-grade voice agent task completion and complex reasoning.
From Google DeepMind Blog
Gemini 3.8 Live for user preference and cost-effectiveness
Choose Gemini 3.8 Live for high user preference and cost-effective voice agent development.
From Google DeepMind Blog
Claude Opus 4.6 for biomedical reference generation
Claude Opus 4.6 offers a good balance of verifiable references and correct author attribution.
From arXiv cs.AI
Claude Opus for biomedical reference generation
Claude Opus 4.8 is the best model for generating biomedical references, with the lowest fabrication rate.
From arXiv cs.AI
GPT Image 2.5 Flare for everyday image generation tasks
Use GPT Image 2.5 Flare for fast, high-quality image generation with text or image inputs.
From Model Documentation
Choose Gemini 3.5 Flash for speed and cost efficiency
Select Gemini 3.5 Flash for tasks requiring high speed, low cost, and agentic capabilities.
From Gemini Models, Gemini Models
Pairing GPT-Live-1 with backend models for tasks
Match reasoning depth, speed, and cost by pairing GPT-Live-1 with Luna or Astra models.
From OpenAI News
Choose between Sunburst and Flare ChatGPT image models
Select the best ChatGPT image model based on your workflow needs.
From Simon Willison
GPT-5.2 Instant for grounded tone and clear answers
Update GPT-5.2 Instant for a more measured tone and clearer, upfront answers to advice questions.
From OpenAI Help Center
GPT-5.3-Codex for agentic coding tasks
Use GPT-5.3-Codex, the most capable agentic coding model, for advanced coding tasks.
From OpenAI Help Center
GPT-5.3 Instant for conversational flow and relevance
Update GPT-5.3 Instant for more accurate answers and better conversational flow.
From OpenAI Help Center
GPT-5.4 mini for fallback reasoning in ChatGPT
GPT-5.4 mini provides fallback reasoning for GPT-5.4 Thinking when rate limits are reached.
From OpenAI Help Center
GPT-5.4 Thinking for upfront planning and context
Use GPT-5.4 Thinking for upfront planning, mid-response adjustments, and improved deep web research.
From OpenAI Help Center
GPT-5.5 Instant for conversational style and quality
Update GPT-5.5 Instant for improved readability, natural conversation, and better pacing in help tasks.
From OpenAI Help Center
GPT-5.6 Sol for complex work in ChatGPT
Use GPT-5.6 Sol for advanced reasoning in coding, research, science, cybersecurity, and design.
From OpenAI Help Center, OpenAI Help Center
Choose GPT-5.6 Luna for free users and GPT-5.6 Sol for paid plans
Free and Go users get GPT-5.6 Luna by default; paid plans can access GPT-5.6 Sol for more advanced reasoning.
From OpenAI Help Center
Free tier model access for ChatGPT users
Free tier users can access GPT-5.6 Luna.
From OpenAI Help Center
Select GPT-6 Pro for top-tier tasks on specific ChatGPT plans
GPT-6 Pro, powered by GPT-6 Astra, is available for Pro $100, Pro $200, Business, and Enterprise ChatGPT plans.
From OpenAI Help Center
Use GPT-5.6 Sol Pro for difficult tasks on eligible ChatGPT plans
GPT-5.6 Sol Pro is the highest-capability GPT-5.6 option for difficult tasks and longer-running workflows on paid plans.
From OpenAI Help Center
Top LLMs for protein design tool use in ABLE benchmark
Claude Sonnet 4 and Gemini 3 Pro show highest scores in LLM agent protein design tasks.
From arXiv cs.AI
GPT-5.5 for agentic coding and knowledge work
Use GPT-5.5 for complex tasks requiring planning, tool use, and iterative action.
From OpenAI Research
HydraFusion selects best model for task
HydraFusion dynamically chooses the most suitable model or model combination for each specific coding task.
From GitHub Blog: Copilot
GPT-5.6 Luna for Cost Sensitive High Volume Workloads
Use GPT-5.6 Luna for demanding tasks where cost and volume are primary concerns.
From Model Documentation
GPT-5.6 Sol for general purpose tasks
GPT-5.6 Sol is the flagship model in the GPT-5.6 family, suitable for general tasks.
From Model Documentation
GPT-6 Astra for complex reasoning and coding tasks
Use GPT-6 Astra for the hardest end-to-end work including complex reasoning, coding, and research.
From Model Documentation
Choose gpt-4o-mini-transcribe for reliable transcription in noisy settings
Use gpt-4o-mini-transcribe for accurate transcription, especially in real-world conversational settings with noise.
From OpenAI Developers Blog
Choose gpt-realtime-mini for low cost and latency voice agents
Use gpt-realtime-mini when cost and latency are critical for voice agents.
From OpenAI Developers Blog
Select gpt-4o-mini-tts for natural sounding text-to-speech
Use gpt-4o-mini-tts for text-to-speech with natural voices and better custom voice consistency.
From OpenAI Developers Blog
Use gpt-realtime for highest accuracy in voice agents
Choose gpt-realtime when agent accuracy is the top priority for voice applications.
From OpenAI Developers Blog
Choose Claude Fable for most capable widely released model
Use Claude Fable 5.1 as Anthropic's most capable widely released model for knowledge work and research.
From Model Overview
Choose Claude Haiku for speed and cost-effectiveness
Use Claude Haiku for quick iteration, lower development costs, and common applications.
From Model Overview
Choose Claude Opus for complex agentic coding and enterprise work
Use Claude Opus for complex agentic coding, deep reasoning, and long-horizon tasks.
From Model Overview
Claude 5.1 for complex AI game development
Use Claude 5.1 for building complex AI-driven games with low to medium effort.
From Wes Roth
Claude Fable 5.1 for agentic tasks
Use Claude Fable 5.1 for long-horizon agentic tasks due to cost and token efficiency.
From Wes Roth
Claude Fable 5.1 for complex, long-running tasks
Use Claude Fable 5.1 for intricate, multi-step tasks and scientific research.
From Anthropic on YouTube, Anthropic on YouTube
Top performing models for scientific claim validation
Claude Opus, Gemma, GPT-5.5, and Claude Fable showed strong performance in scientific claim verification tasks.
From arXiv cs.AI
Ling-RCCA-Flash model for web application generation
Ling-RCCA-Flash model achieves high scores on MiniAppBench and ArtifactsBench for web application generation.
From arXiv cs.AI
Z.ai's GLM-5.3-Flash for low-cost, high-performance AI
GLM-5.3-Flash offers open weights, low pricing, and strong performance, running on domestic chips.
From The Rundown AI
Newer Claude models show improved robotics control capabilities
Recent Claude models demonstrate stronger adaptation and strategy adjustment for robotics tasks.
From Anthropic Research
GPT-5.6 Sol for finance workflows needing editable outputs
GPT-5.6 Sol is recommended for finance workflows requiring editable PowerPoint and Excel outputs with high completion rates.
From OpenAI News
Moonshot AI releases Qimi K3
Qimi K3 is a 2.8T-parameter open-weight model for advanced reasoning.
From Last Week in AI
Anthropic Opus 5 for general tasks at lower cost
Use Opus 5 for agentic coding, knowledge work, and search tasks when Fable 5 is too expensive.
From The Rundown AI
Claude Opus 5 for alignment and safety
Opus 5 is Anthropic's most aligned and safest model to date, adhering better to its Constitution.
From Anthropic News
Claude Opus 5 for coding and knowledge work
Claude Opus 5 is the new state-of-the-art for coding and knowledge work evaluations.
From Anthropic News
Claude Opus 5 for daily use and default model
Opus 5 is designed for everyday use, offering efficiency and strong performance as the default model.
From Anthropic News
Claude Opus 5 for improved visual outputs
Claude Opus 5 offers significantly stronger visual output capabilities compared to previous models.
From Anthropic News
Claude Opus 5 for scientific research
Opus 5 is the most capable generally available model for scientific research, especially in life sciences.
From Anthropic News
Claude Opus 4.7 for advanced software engineering tasks
Use Claude Opus 4.7 for challenging coding work that previously required close supervision.
From Anthropic News
Claude Opus 4.5 for coding and agents
Use Claude Opus 4.5 for coding, agents, and computer use.
From Anthropic News
Claude Sonnet 4.6 for coding and office tasks
Use Sonnet 4.6 for coding, computer use, and office tasks, as it approaches Opus-level intelligence at a lower cost.
From Anthropic News
Claude Opus 4.7 for financial tasks
Claude Opus 4.7 is recommended for state-of-the-art performance on financial tasks.
From Anthropic News
GPT-5.6 models for different tasks and thinking levels
Choose between Luna, Terra, and Sol models with five thinking levels plus Ultra mode for specific tasks.
From Ben's Bites
GPT-5.6 for Microsoft 365 Copilot knowledge work
GPT-5.6 is the preferred model for Microsoft 365 Copilot due to its advanced reasoning for complex knowledge work.
From Microsoft 365 Copilot Blog
Use Claude Opus 5 for complex multi-step work in Microsoft 365 Copilot
Claude Opus 5 is now available in Microsoft 365 Copilot for complex, multi-step tasks.
From Microsoft 365 Copilot Blog, Microsoft 365 Copilot Blog
Use grok-4.20-multi-agent for multi-agent research tasks
Specify grok-4.20-multi-agent for API requests to orchestrate specialized AI agents for deep research.
From xAI Docs: Guides
Choose Gemini Omni Flash for video generation and editing
Use Gemini Omni Flash for high-quality video generation and conversational editing tasks.
From Google DeepMind Blog
Choose Nano Banana 2 Lite for speed and cost in image generation
Use Nano Banana 2 Lite when speed and cost are primary constraints for image generation.
From Google DeepMind Blog
Choose Claude Fable 5 for longer, multi-step workflows in Copilot
Select Claude Fable 5 in Microsoft 365 Copilot for complex tasks needing context and iteration.
From Microsoft 365 Copilot Blog
Choose GPT-5.5 for enterprise agent document parsing
Use GPT-5.5 for agent workflows requiring high accuracy in parsing legacy and scanned documents.
From OpenAI News
GPT-5.5 for creative knowledge work and research
GPT-5.5 is a powerful model for creative knowledge work and research, especially when integrated into tools like Codex.
From OpenAI News
GPT-5.5 for complex coding and knowledge work
Use GPT-5.5 for tasks requiring planning, tool use, and multi-step reasoning like coding and research.
From OpenAI News
GPT-4.1 mini vs. GPT-5 for specific AI tasks
Use GPT-4.1 mini for deterministic tasks, GPT-5 for ambiguous interpretation and reasoning.
From OpenAI News
Use GPT-4.1 for low-latency tool use, GPT-5.2 for planning
Pair GPT-4.1 for fast tool use with GPT-5.2 for deeper, multi-step planning in agentic systems.
From OpenAI News
Claude Opus 4.1 for strong human judgment correlation
Claude Opus 4.1 showed the strongest correlation with human judgment in Bloom evaluation tests.
From Anthropic Research
GPT-5.2-Codex for complex agentic coding tasks
Use GPT-5.2-Codex for advanced agentic coding, real-world software engineering, and project-scale tasks.
From OpenAI News
GPT-5.2 Thinking for professional knowledge work
GPT-5.2 Thinking is the best model for real-world professional use, outperforming experts on knowledge tasks.
From OpenAI News
GPT-5.1 for agentic and coding tasks
GPT-5.1 balances intelligence and speed for agentic and coding tasks, adapting thinking time to task complexity.
From OpenAI News
GPT-5.1 Auto for automatic model selection
Utilize GPT-5.1 Auto for automatic routing of queries to the best-suited model.
From OpenAI News
GPT-5.1 Instant for conversational tasks
Use GPT-5.1 Instant for more conversational interactions and improved instruction following.
From OpenAI News
GPT-5.1 Thinking for precise reasoning tasks
Employ GPT-5.1 Thinking when precise reasoning and adaptive thinking time are crucial.
From OpenAI News
MedGemma for medical text and image generation tasks
Use MedGemma for tasks like report generation or visual question answering involving medical text or images.
From Google DeepMind Blog
MedSigLIP for medical image classification and retrieval
Use MedSigLIP for medical imaging tasks requiring structured outputs like classification or retrieval.
From Google DeepMind Blog
Claude 3.7 Sonnet for coding and front-end development
Claude 3.7 Sonnet is Anthropic's most intelligent model, showing strong improvements in coding.
From Anthropic News
Claude Opus 4 for complex coding and agent tasks
Use Claude Opus 4 for world-leading coding, complex reasoning, and advanced AI agent workflows.
From Anthropic News
Claude Sonnet 4 for balanced coding and reasoning
Use Claude Sonnet 4 for a balance of coding performance, reasoning, and efficiency in everyday use cases.
From Anthropic News
GPT-4.1 for complex, regulated domains
GPT-4.1 is recommended for its instruction following and handling of edge cases in regulated fields.
From OpenAI News
GPT-5 for coding and agentic tasks
GPT-5 is OpenAI's best model for coding and agentic tasks, outperforming previous models on benchmarks.
From OpenAI News
GPT-5 for enhanced business AI applications
Use GPT-5 for advanced AI applications in business, offering superior accuracy, speed, and reasoning.
From OpenAI News
GPT-5 sizes for performance, cost, and latency trade-offs
Choose from gpt-5, gpt-5-mini, and gpt-5-nano for flexible performance, cost, and latency.
From OpenAI News
GPT-5 models for specific tasks and developer access
Select GPT-5 models based on task complexity, speed needs, and developer requirements.
From OpenAI News
Open-weight models for agentic workflows
gpt-oss-120b and gpt-oss-20b are suitable for agentic workflows needing instruction following and tool use.
From OpenAI News
Claude 3.5 Haiku for speed and intelligence benchmarks
Claude 3.5 Haiku matches Opus performance at Haiku speed, excelling in coding.
From Anthropic News
Claude 3.5 Sonnet for coding and agentic tasks
Use upgraded Claude 3.5 Sonnet for significant gains in coding and tool use.
From Anthropic News
GPT-4.1 for research and structured outputs in agents
Use GPT-4.1 for agentic research and structured outputs due to its instruction following and large context window.
From OpenAI News
Computer-Using Agent (CUA) for complex sales research tasks
CUA is used for complex research tasks in sales that static scraping cannot handle.
From OpenAI News
GPT-4o for synthesis and reply classification in sales
GPT-4o is chosen for its structured outputs and fluency in sales synthesis and reply classification.
From OpenAI News
OpenAI o3 for early-stage logic in sales workflows
OpenAI o3 is selected for its two- to three-turn reasoning capabilities in early-stage sales logic.
From OpenAI News
OpenAI o3 for complex reasoning and multi-faceted analysis
Use OpenAI o3 for complex queries needing multi-faceted analysis, especially in programming, business, and creative ideation.
From OpenAI News
OpenAI o4-mini for fast, cost-efficient reasoning and high-volume tasks
Use OpenAI o4-mini for fast, cost-efficient reasoning, especially in math, coding, and visual tasks, and for high-volume usage.
From OpenAI News
GPT-4.1 for coding, instruction following, and long context
Use GPT-4.1 for major gains in coding, instruction following, and long context comprehension.
From OpenAI News
GPT-4.1 mini for improved intelligence and reduced cost
Choose GPT-4.1 mini for intelligence evals matching GPT-4o but with lower latency and cost.
From OpenAI News
GPT-4.1 nano for fast, low-cost classification and autocompletion
Use GPT-4.1 nano for tasks needing speed and low cost, like classification or autocompletion.
From OpenAI News
Use Claude Instant 1.2 for faster, lower-cost AI tasks
Claude Instant 1.2 is a faster, lower-priced model for dialogue, summarization, and document comprehension.
From Anthropic News
GPT-4o for agentic AI in customer service
Use GPT-4o for AI agents that lead conversations and drive resolutions autonomously.
From OpenAI News
OpenAI's o1 series for financial analysis reasoning
OpenAI's o1 series models are recommended for precise financial workflows requiring structured, referenceable reasoning.
From OpenAI News
GPT-4o vision and reasoning for GUI interaction
GPT-4o's capabilities are foundational for CUA's advanced GUI interaction and reasoning.
From OpenAI News
Use o1-2024-12-17 for complex tasks and function calling
New reasoning model o1-2024-12-17 excels at multi-step tasks and structured outputs.
From OpenAI News
Best AI agent setup for ML engineering benchmark
OpenAI's o1-preview with AIDE scaffolding shows strong performance on ML engineering tasks.
From OpenAI News
Use o1 for advanced reasoning tasks
OpenAI o1 model excels in competitive programming, math Olympiads, and complex science benchmarks.
From OpenAI News
GPT-4o fine-tuning for custom applications
Fine-tune GPT-4o for higher performance and accuracy in specific use cases.
From OpenAI News
Use gpt-4o-2024-08-06 for 100% structured output reliability
The latest GPT-4o model achieves perfect scores on structured output evaluations with the new feature.
From OpenAI News
Claude 3 Haiku for fast, affordable enterprise workloads
Choose Claude 3 Haiku for speed and cost-efficiency in enterprise tasks, especially with large datasets.
From Anthropic News
GPT-4o for faster, cheaper, multimodal AI tasks
Use GPT-4o for improved speed, lower cost, and combined text, audio, and vision capabilities.
From OpenAI News
Adopt Chat Completions API over older Completions API models
Migrate to the Chat Completions API for better results and future model investments.
From OpenAI News
Use GPT-4 API for advanced capabilities and existing API customers
Access GPT-4 API for 8K context if you are a paying API customer with a history of successful payments.
From OpenAI News
GPT-4 Turbo for advanced tasks with large context
Use GPT-4 Turbo for complex instructions, specific formats, and large documents up to April 2023.
From OpenAI News
Use gpt-3.5-turbo-0613 for function calling and steerability
The gpt-3.5-turbo-0613 model includes function calling and more reliable steerability via the system message.
From OpenAI News
Use gpt-4-0613 for improved function calling and general performance
The gpt-4-0613 model offers significant improvements in function calling and general task performance.
From OpenAI News
Point-E for fast text-to-3D point cloud generation
Use Point-E for rapid 3D point cloud generation from text when speed is key.
From OpenAI News
Nothing matches. Clear a filter or try a shorter word.