| Claude Haiku 5.5 | Anthropic | October 7, 2026 | active | CursorBench 4.0 (high)42.3% |
| Gemini 4 Argon | Google (DeepMind) | September 30, 2026 | preview | APEX-Agents Original (high)82.4% |
| GPT-6.1 Sol | OpenAI | September 29, 2026 | active | APEX-Agents Original (max)60.0% |
| Claude Sonnet 5.5 | Anthropic | September 28, 2026 | active | APEX-Agents Original (max)75.5% |
| Claude Opus 5.5 | Anthropic | September 22, 2026 | active | APEX-Agents Original (max)73.5% |
| GPT-6 Luna | OpenAI | September 22, 2026 | active | APEX-Agents Original (max)44.3% |
| GPT-6 Sol | OpenAI | September 22, 2026 | active | APEX-Agents Original (max)54.3% |
| Grok 4.7 | SpaceXAI | September 21, 2026 | active | APEX-Agents Original (xhigh)54.6% |
| GPT-6 Astra | OpenAI | September 3, 2026 | active | APEX-Agents Original (max)64.7% |
| Gemini 3.8 Flash | Google (DeepMind) | September 2, 2026 | active | APEX-Agents Original (high)64.3% |
| Muse Spark 1.3 | Meta AI (originally Facebook AI Research) | September 2, 2026 | active | APEX-Agents Original (max)47.6% |
| Claude Fable 5.1 | Anthropic | September 1, 2026 | active | APEX-Agents Original (high)59.7% |
| GLM-5.3-Flash | Z.ai | August 26, 2026 | active | APEX-Agents Original (max)52.8% |
| GLM-5.3 | Z.ai | August 14, 2026 | active | APEX-Agents Original (max)56.6% |
| Gemini 3.7 Flash | Google (DeepMind) | August 13, 2026 | active | APEX-Agents Original (high)67.8% |
| Grok 4.6 | SpaceXAI | August 12, 2026 | active | APEX-Agents Original (xhigh)65.3% |
| MAI-Thinking-1 | Microsoft AI | August 12, 2026 | active | Riemann-bench October 2026 leaderboard4.8% |
| Muse Glimmer 30B | Meta AI (originally Facebook AI Research) | August 10, 2026 | active | Chartography October 2026 leaderboard (xHigh)17.2% |
| Muse Spark 1.2 | Meta AI (originally Facebook AI Research) | August 5, 2026 | active | APEX-Agents Original (xhigh)36.4% |
| Qwen3.8-Max | Alibaba (Tongyi) | August 2, 2026 | active | APEX-Agents Original (xhigh)63.3% |
| Inkling-Small | Thinking Machines Lab | July 30, 2026 | active | Chartography October 2026 leaderboard (xHigh)9.1% |
| Claude Opus 5 | Anthropic | July 24, 2026 | active | APEX-Agents Original (max)65.8% |
| Gemini 3.6 Flash | Google (DeepMind) | July 21, 2026 | active | APEX-Agents Original (high)46.9% |
| Grok 4.5 | SpaceXAI | July 16, 2026 | active | APEX-Agents Original (high)56.2% |
| Kimi K3 | Moonshot AI | July 16, 2026 | active | APEX-Agents Original (max)50.6% |
| Inkling | Thinking Machines Lab | July 15, 2026 | active | APEX-Agents Original (high)33.8% |
| GPT-5.6 Luna | OpenAI | July 9, 2026 | active | APEX-Agents Original (max)43.0% |
| GPT-5.6 Sol | OpenAI | July 9, 2026 | active | AutomationBench 1.0.63.8% |
| GPT-5.6 Terra | OpenAI | July 9, 2026 | active | APEX-Agents Original (max)58.2% |
| Muse Spark 1.1 | Meta AI (originally Facebook AI Research) | July 9, 2026 | active | APEX-Agents Original (xhigh)31.8% |
| Claude Sonnet 5 | Anthropic | June 30, 2026 | active | APEX-Agents Original (max)54.5% |
| GLM-5.2 | Z.ai | June 16, 2026 | active | APEX-Agents Original (max)45.2% |
| Claude Fable 5 | Anthropic | June 9, 2026 | active | APEX-Agents Original (max)63.6% |
| NVIDIA Nemotron 3 Ultra 550B-A55B | NVIDIA | June 4, 2026 | active | APEX-Agents Original (high)22.7% |
| Gemma 4 12B Instruct | Google (DeepMind) | June 3, 2026 | active | BIG-Bench Extra Hard Original — micro average (thinking mode)53.0% |
| Qwen3.7-Plus | Alibaba (Tongyi) | June 3, 2026 | active | AutomationBench 1.0.63.7% |
| MAI-Code-1-Flash | Microsoft AI | June 2, 2026 | active | GPQA Diamond84.6% |
| MiniMax M3 | MiniMax | June 1, 2026 | active | APEX-Agents Original (high)37.7% |
| Claude Opus 4.8 | Anthropic | May 28, 2026 | active | APEX-Agents Original (max)48.9% |
| Mistral Medium 3.5 | Mistral AI | May 22, 2026 | active | Finance Agent Benchmark 2 — October 2026 evaluation (high)32.1% |
| Qwen3.7-Max | Alibaba (Tongyi) | May 20, 2026 | active | GPQA Diamond — Epoch AI harness90.9% |
| Gemini 3.5 Flash | Google (DeepMind) | May 19, 2026 | active | APEX-Agents Original (high)27.5% |
| Composer 2.5 | Cursor | May 18, 2026 | active | Vibe Code Bench 1.1 — Cursor CLI, October 2026 evaluation49.6% |
| DeepSeek-V4-Flash | DeepSeek AI | April 24, 2026 | active | AutomationBench 1.0.6 (max)18.1% |
| DeepSeek-V4-Pro | DeepSeek AI | April 24, 2026 | active | BrowseComp 130-question subset — Mercor web research agent (max)62.3% |
| GPT-5.5 | OpenAI | April 23, 2026 | active | APEX-Agents Original (xhigh)55.1% |
| Kimi K2.6 | Moonshot AI | April 20, 2026 | active | AutomationBench 1.0.62.6% |
| Qwen3.6-35B-A3B | Alibaba (Tongyi) | April 17, 2026 | active | GPQA Diamond — Epoch AI harness83.8% |
| Claude Opus 4.7 | Anthropic | April 16, 2026 | active | APEX-Agents Original (max)49.2% |
| Muse Spark | Meta AI (originally Facebook AI Research) | April 8, 2026 | active | GPQA Diamond — Epoch AI harness89.8% |