Coding

💻Coding 🐞Debugging 🤖Agents 🎯Code autocompletion

Grok Build 0.1

By xAI

Grok Build 0.1 is xAI’s fast coding model for agentic coding, web development, debugging, MCP support, and tool-calling workflows.

NewMultimodal

Released 6d ago

💻Coding 🎯Code autocompletion 💻Conversational coding

Mellum2 12B A2.5B Base

By JetBrains

Mellum2-12B-A2.5B-Base is JetBrains’ Apache 2.0 text-generation base model, a 12B MoE checkpoint with 2.5B active parameters for long-context code and language tasks.

NewCoding

Released 6d ago

🚀Productivity 💵Financial analysis 💻Coding 🤖Agents 🔍Image interpretation

Claude Opus 4.8

By Anthropic

Claude Opus 4.8 is Anthropic's new flagship model, released May 28, 2026. It improves on Opus 4.7 with stronger coding, more honest self-assessment, and a faster, cheaper fast mode, at the same standard pricing. New features include user-controlled effort levels and Dynamic Workflows for parallel subagents.

NewMultimodal

Released 7d ago

💻Coding 🐞Debugging 🔧Code refactoring 🔬Test generation

Composer 2.5

By Anysphere

Composer 2.5 is Cursor’s coding-agent model for long-running software tasks, complex instruction following, and agentic coding inside Cursor.

NewMultimodal

Released 16d ago

📚Large Language Models 🔎Problem solving 💻Coding 🔢Math

ZAYA1-8B

By Zyphra AI

ZAYA1-8B is Zyphra’s small mixture-of-experts reasoning model, built for long-form reasoning, mathematics, and coding. It has 8.4B total parameters with 760M active parameters, is post-trained for reasoning, and is positioned as highly intelligence-efficient enough for both strong test-time-compute use and local on-device deployment.

NewMultimodal

Released 29d ago

Gen 7 Grok

Grok 4.3

By xAI

Grok 4.3 is xAI’s flagship large language model for high-end reasoning, coding, and agentic workflows. xAI positions it as its most truth-seeking model, with a 1,000,000-token context window and support across the broader xAI API stack for tools, reasoning, and advanced integrations.

💬Chatting 💻Coding 🤖Agents 💰Finance ⚖️Legal 🤔Logical reasoning

NewText

Released 1mo ago

💬Chatting 💼Business 💻Coding 🤖Agents ⚖️Legal 🔢Math

ERNIE 5.1

By Baidu

ERNIE-5.1 is Baidu’s new preview flagship language model, built for stronger general text capability, better cost efficiency, and improved creative performance. It is positioned as the top-ranked Chinese model on the LMArena Text leaderboard and as a high-efficiency upgrade over ERNIE-5.0 with much lower training cost at its scale.

NewText

Released 1mo ago

💻Coding 🐞Debugging 🤖Agents 🤔Logical reasoning 💻Software engineer assistance

Laguna XS 2

By poolside

Laguna XS.2 is poolside’s open 33B MoE coding model for agentic software work on local machines. It uses 3B activated parameters per token, supports native reasoning with interleaved thinking around tool calls, and is designed for long-horizon coding with low KV-cache requirements and Mac-friendly deployment.

NewMultimodal

Released 1mo ago

💬Chatting 💻Coding 🤖Agents 🤔Logical reasoning 🔍Image interpretation

Mistral Medium 3.5

By Mistral AI

Mistral Medium 3.5 is Mistral’s new flagship merged model for instruction following, reasoning, and coding. It is a dense 128B model with a 256K context window, built for long-horizon productivity and agentic work, with configurable reasoning effort and strong self-hosted efficiency.

NewMultimodal

Released 1mo ago

🤖Agents 🐞Debugging 💻Coding 🔍Image interpretation

Carnice V2 27B

By Kai Stephens

Carnice-V2-27B is a BF16 supervised fine-tune of Qwen3.6-27B for Hermes-style agent traces. It is built for agentic conversational use, instruction following, and tool-oriented workflows, and is released as a fully merged standalone checkpoint rather than only a LoRA adapter.

NewText

Released 1mo ago

💬Chatting 💻Coding 🤖Agents 🎩Sophisticated reasoning

Hy3 preview

By Tencent

Hy3 Preview is Tencent’s new open-source flagship AI model, built for real-world business use and stronger agentic performance. SCMP reports it has 295B parameters, is Tencent’s most powerful model so far, and is already being deployed in products including Yuanbao and CodeBuddy.

NewMultimodal

Released 1mo ago

💻Coding 🔍Code reviews ✉️Codebase Q&A 🤖Agents 🎩Sophisticated reasoning

MiMo V2.5 Pro

By Xiaomi

MiMo-V2.5-Pro is Xiaomi’s latest agentic language model, built for stronger long-horizon coherence, complex software engineering, and autonomous tool-using workflows. It is positioned as Xiaomi’s most capable model so far, with better instruction following in agent settings, stronger ultra-long-context consistency, and support for very long tool-use trajectories.

NewText

Released 1mo ago

Gen 3 Qwen

Qwen 3.6 27B

By Alibaba

Qwen3.6-27B is Qwen’s open-weight multimodal model for coding, agent workflows, long-context reasoning, and vision-language tasks. It combines a 27B causal language model with a vision encoder, supports image-text-to-text use, and offers a native 262,144-token context window extendable to about 1.01M tokens.

💬Chatting 💻Coding 🤖Agents 🔍Image interpretation 🎩Sophisticated reasoning

NewMultimodal

Released 1mo ago

💬Chatting 💻Coding 🤖Agents 📄Document processing

Ling 2.6 flash

By AntGroup

Ling-2.6-Flash is Ant Ling’s high-efficiency instruct language model, built for strong real-agent performance with lower token usage and lower deployment cost. From the launch post in your screenshot, it has 104B total parameters with 7.4B active parameters and is positioned for better throughput, lower cost, and more efficient large-scale deployment

NewText

Released 1mo ago

Gen 3 Qwen

Qwen 3.6 Max Preview

By Alibaba

Qwen 3.6 Max Preview appears to be Alibaba’s newest preview flagship in the Qwen line, positioned above Qwen3.6-Plus and aimed at top-end coding, world knowledge, and instruction-following performance. I could not read the exact official blog page you linked directly, so I’m relying on current public reporting for this description.

💬Chatting 💻Coding 🤖Agents 🎩Sophisticated reasoning

NewMultimodal

Released 1mo ago

Gen 3 Kimi

Kimi K2.6

By Moonshot AI

Kimi-K2.6 is Moonshot AI’s open-source native multimodal agentic model, built for long-horizon coding, coding-driven design, proactive autonomous execution, and large-scale multi-agent orchestration. It uses a MoE architecture with 1T total parameters, 32B active parameters, a 256K context window, and a MoonViT vision encoder.

💬Chatting 💻Coding 🤖Agents 🔍Image interpretation 🎩Sophisticated reasoning

NewMultimodal

Released 1mo ago

💬Chatting 💵Financial analysis 💻Coding 🤖Agents 🔍Image interpretation

Claude Opus 4.7

By Anthropic

Claude Opus 4.7 is Anthropic’s latest generally available frontier model, tuned for advanced software engineering, long-running autonomous tasks, stronger instruction following, and better high-resolution vision. It is positioned as a clear upgrade over Opus 4.6, especially for difficult coding work, while keeping the same pricing.

NewMultimodal

Released 1mo ago

💻Coding 🤖Agents 🔍Image interpretation 🔢Math 💻Conversational coding

Qwen 3.6 35B A3B

By Alibaba

Qwen3.6-35B-A3B is Qwen’s open-weight multimodal MoE model for coding, agentic workflows, long-context reasoning, and vision-language tasks. It has 35B total parameters with 3B activated, supports image-text-to-text use, preserves reasoning context across turns, and natively handles 262,144 tokens with extension up to about 1.01M.

NewMultimodal

Released 1mo ago

💻Coding 🤖Task automation 🔄Workflow optimization

Windsurf 2.0

By Cognition AI

Windsurf 2.0 is Cognition’s upgraded agentic coding workspace centered on local and cloud agents working together. It introduces an Agent Command Center for managing parallel work across repos and environments, and adds Devin integration inside Windsurf, aiming to make software development more orchestration-driven and less tied to a single editor session.

NewText

Released 1mo ago

💻Coding 🤖Task automation 💻Conversational coding

GLM 5.1

By Z.ai

GLM-5.1 appears to be Z.ai’s coding-focused update to GLM-5, aimed at stronger agentic software engineering and long-context coding. Public reporting describes it as a post-training upgrade rather than a new base architecture, keeping the large MoE backbone while improving coding performance.

NewText

Released 1mo ago

💻Coding 🔍Code reviews 🐞Debugging

SWE 1.6

By Cognition AI

SWE-1.6 is Cognition’s latest model for software engineering agents, built to improve both raw capability and what the company calls “model UX.” It is positioned as fast, low-cost, and smoother to use in practice, with up to 950 tok/s and better behavior in agent workflows.

NewText

Released 1mo ago

💻Coding 🔧Code optimization 🏗️Software architecture guidance 🔌Api integration assistance

KAT Coder Pro V2

By Kuaishou Technology

KAT-Coder-Pro V2 is a hosted Kwaipilot coding model listed on OpenRouter with a 256,000-token context window. It is positioned as a large-context software engineering model for coding and agent-style development tasks, intended for API-based use rather than open-weight local deployment.

NewMultimodal

Released 2mo ago

💻Coding 🔧Code refactoring 💻Vibe coding

Composer 2

By Anysphere

Composer 2 is Cursor’s frontier coding model optimized for high intelligence per dollar, built to solve long-horizon software engineering tasks with many tool actions.

NewCoding

Released 2mo ago

Gen 3 GPT

GPT 5.4

By OpenAI

GPT-5.4 is OpenAI’s closed-source frontier model for professional work, focused on coding, reasoning, computer use, tool use, and long-context workflows. It supports up to about 1M tokens of context and is available through ChatGPT, API, and Codex.

💬Chatting 💻Coding 📷Images

NewMultimodal

Released 2mo ago

🖼️Image generation 💻Coding 📝Writing 📷Images

Qwen 3.5 35B A3B

By Alibaba

wen3.5-35B-A3B is a Mixture-of-Experts Qwen3.5 model with 35B total parameters and 3B active per token, built for strong capability with much lower active compute.

Multimodal

Released 3mo ago

💻Coding 🔍Code reviews ✉️Codebase Q&A 🔧Code refactoring 👨‍💻Collaborative coding

Kilo

By Kilo Code

Kilo is an agentic engineering platform and open-source coding agent that can generate code from natural language, run terminal commands, and automate browser-based steps, available as a VS Code extension and a CLI.

Multimodal

Released 3mo ago

Gen 2 Qwen

Qwen 2.5 Coder 32B

By Alibaba

Code-focused Qwen model family aimed at code generation, reasoning, and fixing across multiple parameter sizes.

Coding

Released 3mo ago

💻Coding 🤖Agents 🖍Coloring books

Claude Sonnet 4.6

By Anthropic

Anthropic’s Sonnet 4.6 model for scaled production and complex tasks across coding, agents, and professional workflows.

Text

Released 3mo ago

Gen 3 Grok

Grok 4.2

By xAI

Grok 4.2 is referenced as a Grok 4-series update for xAI’s Grok assistant, focused on general-purpose help across reasoning, coding, and real-time info retrieval inside the Grok experience.

📷Images 💻Coding 📚Learning 👗Fashion

Multimodal

Released 3mo ago

Gen 3 Qwen

Qwen3.5 397B A17B

By Alibaba

Qwen3.5-397B-A17B is a native vision-language model from Alibaba's Qwen team with 397B parameters (17B active per token). In the Qwen3.5-Plus hosted variant it supports up to 1M-token multimodal context, with strong reasoning, coding, and agentic tool use for large-scale deployment.

🖼️Image generation 💻Coding 📷Images 💬Chatting

Multimodal

Released 3mo ago

GPT 5.3 Codex Spark

By OpenAI

GPT-5.3-Codex-Spark is OpenAI’s first real-time coding model, a smaller GPT-5.3-Codex variant that streams over 1,000 tokens per second with a 128k context window on Cerebras hardware, in research preview for ChatGPT Pro users.

💻Coding 🎭Face swapping 💡Startup advice 📝Content

Coding

Released 3mo ago

Gen 7 GLM

GLM 5

By Z.ai

GLM 5 is Z.ai's 744B parameter open Mixture-of-Experts flagship model for agentic engineering, delivering frontier level coding, reasoning and long horizon agent performance while remaining Apache 2.0 licensed and optimized for Chinese and global hardware.

Text

Released 3mo ago

Gen 2 GPT

GPT 5.2 Codex

By OpenAI

GPT-5.2-Codex is OpenAI’s frontier agentic coding model based on GPT-5.2, optimized for long-horizon software work, large refactors and migrations, better Windows support and state-of-the-art cybersecurity performance in Codex.

💻Coding 🎭Face swapping 🇯🇵Japanese lessons 💻Coding guidance

Coding

Released 3mo ago

Gen 2 GPT

GPT 5.3 Codex

By OpenAI

GPT-5.3-Codex is OpenAI’s most capable agentic coding model, combining GPT-5.2-Codex’s frontier coding with GPT-5.2’s reasoning in one model that is ~25% faster, tops SWE-Bench Pro and Terminal-Bench, and runs long, tool-using workflows on a computer.

💻Coding 🔍Code reviews 🔎Problem solving 📝Writing

Coding

Released 3mo ago

🎥Storyboards 💻Coding 🚀Productivity 🔍Research

Claude Opus 4.6

By Anthropic

Claude Opus 4.6 is Anthropic’s newest frontier model, built for complex reasoning, large-scale coding and multi-agent orchestration, with major gains in cybersecurity bug-finding and long-horizon enterprise workflows over Opus 4.5.

Multimodal

Released 3mo ago

Gen 2 Qwen

Qwen Code Qoder

By Alibaba

Qwen-Coder-Qoder is a reinforced code model based on Qwen-Coder, custom trained for the Qoder agentic coding platform to improve end to end programming performance inside its IDE and CLI workflows

💻Coding 🔍Code reviews 🔧Code refactoring 🎯Code autocompletion 💻Software engineering guidance

Coding

Released 4mo ago

Gen 2 Qwen

Qwen 3 Coder Next

By Alibaba

Qwen3-Coder-Next is an open-weight language model built specifically for coding agents and local development, optimized for tool use and long-horizon software engineering tasks.

Coding

Released 4mo ago

Gen 7 Step

Step 3.5 Flash

By StepFun

Step 3.5 Flash is StepFun’s open source sparse MoE LLM, with 196B total parameters but 11B active per token, tuned for fast agentic reasoning, coding and long context work while staying efficient.

💻Coding 📝Writing 🌪️Brainstorming 🤖Agents

Text

Released 4mo ago

Gen 7 Trinity

Trinity Large Preview

By Arcee AI

Trinity Large Preview is a 400B-parameter (about 13B active) frontier Trinity MoE model with long-context comprehension, tuned for strong reasoning, coding and multi-step agents, served via hosted preview APIs.

💻Coding 🇯🇵Japanese lessons 🔒Cybersecurity education 🧪React native testing

Text

Released 4mo ago

💻Coding 🔎Problem solving 🇯🇵Japanese lessons

SERA 32B GA

By Ai2

SERA-32B-GA is a 32B Qwen 3 based coding agent variant trained on GLM-4.5-Air trajectories, delivering about 46.6% SWE-bench Verified performance as a slightly smaller-teacher alternative to SERA-32B.

Text

Released 4mo ago

💻Coding 📚Code explanations 🔍Code reviews 💡Coding help

SERA 32B

By Ai2

SERA-32B is Ai2’s flagship 32B Open Coding Agent, a Qwen 3-32B based repository-level coding model trained with Soft Verified Generation that reaches about 49.5% on SWE-bench Verified, rivaling much larger proprietary systems.

Text

Released 4mo ago

SERA 8B GA

By Ai2

SERA-8B-GA is an 8B Qwen 3 based coding agent variant trained from GLM-4.5-Air generated trajectories, matching SERA-8B’s 31.7% SWE-bench Verified score while providing an alternative teacher setup for research.

Text

Released 4mo ago

💻Coding 📚Code explanations 🔍Code reviews

SERA-8B

By Ai2

SERA-8B is Ai2’s 8B-parameter open-source coding agent, fine-tuned from Qwen 3-8B with Soft Verified Generation to solve repository-level bugs, reaching about 31.7% resolve rate on SWE-bench Verified at 32K context.

Text

Released 4mo ago

Gen 3 Kimi

Kimi K2.5

By Moonshot AI

Kimi K2.5 is Moonshot’s native multimodal open model, trained on about 15T vision-text tokens to power advanced coding and visual reasoning and to orchestrate self-directed agent swarms with up to 100 sub-agents and 1,500 tool calls.

💻Coding 📝Writing 📷Images 🎥Videos

Multimodal

Released 4mo ago

Gen 7 Qwen

Qwen3 Max Instruct

By Alibaba

Qwen3 Max Instruct is the trillion-parameter chat and tool-use version of Qwen3-Max, ranking near the top of LMArena and scoring about 69.6 on SWE-Bench Verified and 74.8 on Tau2 Bench for real-world coding and agent tasks.

💻Coding 💬Chatting 🤖Agents 🚀Productivity

Text

Released 4mo ago

Gen 7 GLM

GLM 4.7 Flash

By Z.ai

GLM-4.7-Flash is ZhipuAI’s 30B-class MoE reasoning and coding model with about 3B active parameters, a 200K context window, and strong tool-calling, tuned as a high-speed, lightweight variant of GLM-4.7 for local deployment and agentic coding.

💻Coding 🎨NFT art 📚Stories

Text

Released 4mo ago

💻Coding 💡Coding help 🎭Face swapping

NousCoder 14B

By Nous Research

NousCoder-14B is a 14B-parameter coding model post-trained from Qwen3-14B with RL, tuned for competitive Olympiad-style programming and reaching about 67.9 percent Pass@1 on LiveCodeBench v6.

Text

Released 4mo ago

Gen 7 LFM

LFM2.5 1.2B

By Liquid AI

LFM2.5-1.2B is Liquid AI’s 1.2B hybrid edge model for on-device chat and tools, trained on 28T tokens and RL, running under 1 GB of RAM while rivaling much larger LLMs on reasoning and agentic tasks.

💬Chatting 💻Coding 🤖Chatbots 🚀Productivity

Text

Released 4mo ago