Every model on the Opper gateway

Anthropic's Opus model for complex agentic coding and enterprise work. 1M context, adaptive thinking.

Claude Fable 5 by Anthropic, state-of-the-art generalist model with advanced vision and 1M token context for complex software engineering and reasoning.

OpenAI's frontier GPT-5.6 model for complex professional work. 1M context window with reasoning.

Kimi K3 by Moonshot: open-weight 2.8 trillion parameter MoE with 104 billion active, native vision, and a 1M token context for long-horizon coding and agentic work.

Claude Opus 4.8 by Anthropic, highly capable model for complex reasoning, coding, and agentic work with 1M context and improved efficiency.

GPT-5.6 model that balances intelligence and cost. 1M context window with reasoning.

GPT-5.5 is OpenAI's newest frontier model, combining strong coding, reasoning, and web search and understanding multi-part tasks, with a 1M context window and vision.

xAI's Grok 4.5 model with toggleable reasoning, 500K context window, vision, and agentic tool use

Claude Opus 4.7 by Anthropic, advanced vision and coding model with 1M context for complex long-running agentic workflows.

Latest Sonnet model, balanced for most tasks

GPT-5.4 is OpenAI's frontier model for complex professional work, with a 1M context window, vision, tool use, structured output, and configurable reasoning.

GPT-5.6 model optimized for cost-sensitive, high-volume workloads. 1M context window with reasoning.

GLM-5.2 by Z.ai is a 744B open-weight (MIT) mixture-of-experts coding model with a 1M-token context, IndexShare sparse attention, and selectable thinking-effort levels.

Gemini 3.5 Flash by Google pairs frontier-level intelligence with fast throughput and a 1M token context for agentic and coding tasks.

Google Gemini 3.6 Flash — improved token efficiency and agentic planning at a lower price point than 3.5 Flash

DeepSeek V4 Flash 0731, the July 2026 retrained checkpoint with major agentic and coding gains over the original release.

Also available as deepseek-v4-flash-latest.

DeepSeek V4 Flash, lightweight 284B MoE with 1M context and hybrid thinking for cost-efficient agents.

Claude Sonnet 4.6 by Anthropic, full upgrade across coding, computer use, and long-context reasoning.

Gemini 3.1 Pro Preview by Google, a frontier reasoning LLM with 1M context and strong agentic coding, scoring 77% on ARC-AGI-2 and 81% on SWE-Bench Verified.

Qwen3.7-Max by Alibaba, a 1M-context proprietary reasoning agent with extended thinking for multi-step code and autonomous workflows.

MiniMax M3: natively multimodal 428B Mixture-of-Experts model with ~23B active parameters, MiniMax Sparse Attention for long context, and 59.0% on SWE-Bench Pro

DeepSeek V4 Pro, flagship 1.6T MoE with 1M context and hybrid thinking for frontier reasoning and agents.

GPT-5.3 Codex is OpenAI's most capable agentic coding model, reaching state-of-the-art on SWE-Bench Pro and Terminal-Bench while using fewer tokens than prior models.

Kimi K2.6 by Moonshot: open-weight agentic model with long-horizon coding, 300-agent swarms, video understanding, and 256K context.

Claude Opus 4.6 by Anthropic, professional knowledge-work model with 1M context, extended thinking, and 128K token outputs.

GPT-5.2 by OpenAI, flagship coding and agentic model with a 272K context, vision, reasoning, and tool use.

Kimi K2.7 Code by Moonshot: open-weight, code-specialized agentic model on a 1 trillion parameter MoE, tuned for long-horizon software engineering and tool use.

Claude Opus 4.5 by Anthropic, highly capable model for coding and agents with extended thinking and a tunable effort parameter.

GLM-5.1 by Z.ai is a 754B open-weight (MIT) coding and agentic reasoning model with 202K context that tops SWE-Bench Pro at 58.4.

GPT-5.4 Mini is OpenAI's strongest small model, improving over GPT-5 Mini in coding and reasoning while running more than 2x faster, with a 400k context window.

GLM 5 by Zhipu: 744B frontier MoE model with 40B active, reasoning-first design, and Huawei Ascend training.

Qwen3.7-Plus by Alibaba, a 1M-context multimodal agent with vision, tool calling, and agentic iteration for GUI and CLI automation.

GPT-5.4 Nano is the fastest, lightest GPT-5.4 variant, built for classification, data extraction, and lightweight coding subagents that need low latency.

MiniMax M2.7: self-evolving 229B model with 56.22% SWE-Pro, native multi-agent teams, and 1495 ELO on GDPval-AA (highest open-weight)

MiniMax M2.7-highspeed: throughput-optimized M2.7 variant delivering the self-evolving model's capabilities at higher output speed with Agent Teams

GLM-5-Turbo by Z.ai is a speed-optimized agentic model built for OpenClaw workflows, with thinking modes, reliable tool calling, and 200K context.

Grok 4.3 by xAI combines always-on reasoning with a 1M context window, native vision, and agentic tool calling in one model.

Qwen3.6-27B by Alibaba: 27B open-weight multimodal model with strong agentic coding (77.2 SWE-bench Verified) and reasoning across 262K context.

GPT-5.1 by OpenAI, reasoning model with a 272K-token context, vision, and structured output for complex problem-solving.

Google Gemini 3.5 Flash Lite — low-latency, highly cost-effective subagent model for high-volume automation

Claude Sonnet 4.5 by Anthropic, strong coding and agentic model scoring 61.4% on the OSWorld computer-use benchmark, 30+ hour focus.

Kimi K2.5 by Moonshot: native multimodal 1T-parameter MoE with vision, 256K context, hybrid reasoning, and an agent swarm for parallel task execution.

GPT-5 by OpenAI, multimodal reasoning model with a 272K-token context window for coding, analysis, and agentic work.

GLM-5V-Turbo by Z.ai is a native multimodal model with vision, text, and video input that turns design mockups into front-end code.

Qwen3.5-27B by Alibaba: 27B multimodal model with vision, tool calling, and thinking mode for reasoning and coding.

Qwen3.5 397B A17B by Alibaba: large sparse MoE model with 397B total (17B active) parameters for advanced reasoning and coding.

Claude Opus 4.1 by Anthropic, improved agentic reasoning and multi-file code refactoring at 74.5% SWE-bench.

MiniMax M2.5: 229B MoE model trained on hundreds of thousands of real environments, scoring 80.2% on SWE-Bench Verified and 76.3% on BrowseComp