| Muse Wonder: A Frontier Model For Personal & Professional AI Agents | | 0.79 | Muse Wonder Full (max) (MiniSWE-Agent) | 2026-10-12 |
| Muse Wonder: A Frontier Model For Personal & Professional AI Agents | | 0.788 | Muse Wonder (max) (MiniSWE-Agent) | 2026-10-12 |
| Gemini 4 Argon: our next era of frontier intelligence | ✓ Link | 0.779 | Gemini 4 Argon (mini-swe-agent) | 2026-09-30 |
| DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression | ✓ Link | 0.77 | DeepSeek-V4.1-Flash-1017 (Mini-SWE-V2) | 2026-09-17 |
| Introducing Muse Spark 1.3 | | 0.754 | Muse Spark 1.3 (max effort) (mini-swe-agent) | 2026-09-02 |
| Introducing GPT-6.1 Sol | | 0.7522 | GPT-6.1 Sol (high effort) (mini-swe-agent) | 2026-09-29 |
| DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression | ✓ Link | 0.742 | DeepSeek-V4.1-Flash Max (mini-SWE) | 2026-09-17 |
| Introducing Claude Opus 5.5 | ✓ Link | 0.742 | Claude Opus 5.5 | 2026-09-22 |
| GPT-6 Astra: A new generation of intelligence | ✓ Link | 0.741 | GPT-6 Astra | 2026-09-03 |
| Introducing Gemini 3.8 Flash and 3.8 Flash Cyber | ✓ Link | 0.737 | Gemini 3.8 Flash (high thinking) (mini-swe-agent) | 2026-09-02 |
| Introducing SWE-2: Pushing the Pareto Frontier | | 0.73 | SWE-2 (mini-swe-agent) | 2026-09-10 |
| GPT-5.6: Frontier intelligence that scales with your ambition | ✓ Link | 0.727 | GPT-5.6 Sol | 2026-07-09 |
| MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement | ✓ Link | 0.719 | MiMo-V2.6-Pro-RL (mini-swe-agent) | 2026-10-08 |
| Introducing Grok 4.7 | | 0.71 | Grok 4.7 (high effort) | 2026-09-21 |
| Introducing Claude Sonnet 5.5 | ✓ Link | 0.71 | Claude Sonnet 5.5 (max effort) | 2026-09-28 |
| DeepSWE Leaderboard | ✓ Link | 0.6991 | Claude Fable 5 (xhigh) (MiniSWE-Agent) | 2026-06-09 |
| Introducing GPT-6 Sol and Luna | | 0.6881 | GPT-6 Sol (max effort) (mini-swe-agent) | 2026-09-22 |
| Introducing Claude Opus 5 | ✓ Link | 0.688 | Claude Opus 5 | 2026-07-24 |
| Step 5 Preview | | 0.677 | Step 5 Preview (High) (SWE-agent) | 2026-09-20 |
| Kimi K3: Open Frontier Intelligence | ✓ Link | 0.675 | Kimi K3 (Max) (Kimi Code) | 2026-07-27 |
| DeepSWE Leaderboard | | 0.6748 | Grok 4.6 (medium) (MiniSWE-Agent) | 2026-08-12 |
| Introducing Claude Fable 5.1 and Claude Mythos 5.1 | | 0.674 | Claude Fable 5.1 (max effort) | 2026-09-01 |
| DeepSWE Leaderboard | | 0.6704 | GPT-5.5 (xhigh) (MiniSWE-Agent) | 2026-04-23 |
| GLM-5.3: Frontier Coding with Emergent Cyber Capabilities | ✓ Link | 0.669 | GLM-5.3 (MiniSWE-Agent) | 2026-08-14 |
| Introducing GPT-6 Sol and Luna | | 0.6659 | GPT-6 Luna (max effort) (mini-swe-agent) | 2026-09-22 |
| Introducing Gemini 3.7 Flash | ✓ Link | 0.653 | Gemini 3.7 Flash (High) (MiniSWE-Agent) | 2026-08-13 |
| Hy4 preview | ✓ Link | 0.643 | Hy4 preview (mini-swe-agent) | 2026-08-28 |
| GLM-5.3-Flash: Frontier Intelligence, Flash Cost | ✓ Link | 0.634 | GLM-5.3-Flash (mini-swe-agent) | 2026-08-26 |
| DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence | ✓ Link | 0.627 | DeepSeek-V4-Pro-0813 (Max) (DeepSeek Harness) | 2026-04-26 |
| Mistral Large 4 | | 0.617 | Mistral Large 4 Preview | 2026-10-06 |
| On the Limitations of Non-GPU AI Accelerators for Large-Model Inference: A Field Study of MoE and Multimodal Serving on Huawei Ascend | | 0.593 | DeepSeek-V4-Flash-Vision-Exp (DeepSeek Harness) | 2026-07-09 |
| DeepSWE Leaderboard | | 0.5897 | Claude Opus 4.8 (max) (MiniSWE-Agent) | 2026-05-28 |
| On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability | ✓ Link | 0.587 | Qwen3.8-Flash-Next (mini-SWE-agent) | 2026-08-31 |
| Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA | ✓ Link | 0.584 | Macaron-V1-Venti (Claude Code) | 2026-08-10 |
| DeepSWE Leaderboard | ✓ Link | 0.5746 | Qwen3.8 Max (xhigh) (MiniSWE-Agent) | 2026-08-03 |
| DeepSWE Leaderboard | | 0.5487 | Muse Spark 1.2 (xhigh) (MiniSWE-Agent) | 2026-08-05 |
| DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence | ✓ Link | 0.544 | DeepSeek-V4-Flash-0731 (Max) (DeepSeek Harness) | 2026-04-26 |
| DeepSWE Leaderboard | | 0.5385 | Claude Sonnet 5 (max) (MiniSWE-Agent) | 2026-06-30 |
| Introducing Muse Spark 1.1 | ✓ Link | 0.533 | Muse Spark 1.1 (MiniSWE-Agent) | 2026-07-09 |
| Introducing Grok 4.5 | | 0.53 | Grok 4.5 | 2026-07-08 |
| DeepSWE Leaderboard | | 0.5177 | GPT-5.4 (xhigh) (MiniSWE-Agent) | 2026-03-05 |
| DeepSWE Leaderboard | | 0.4668 | Gemini 3.6 Flash (high) (MiniSWE-Agent) | 2026-07-21 |
| GLM-5.2: Built for Long-Horizon Tasks | ✓ Link | 0.462 | GLM-5.2 (MiniSWE-Agent) | 2026-06-16 |
| Introducing Beam: Reflection's 501B open-weight model | | 0.444 | Beam | 2026-10-05 |
| Qwen3.8-27B | ✓ Link | 0.422 | Qwen3.8-27B (Claude Code) | 2026-08-05 |
| Introducing Laguna S 2.1 | ✓ Link | 0.404 | Laguna S 2.1 (pool) | 2026-07-21 |
| DeepSWE Leaderboard | | 0.3606 | Gemini 3.5 Flash (high) (MiniSWE-Agent) | 2026-05-19 |
| Seed2.1 Model Card: Agentic Intelligence for Productivity | | 0.327 | Seed2.1 Pro | 2026-06-23 |
| DeepSWE Leaderboard | | 0.3053 | Kimi K2.7 Code (MiniSWE-Agent) | 2026-06-11 |
| DeepSWE Leaderboard | | 0.2993 | Claude Sonnet 4.6 (high) (MiniSWE-Agent) | 2026-02-17 |
| Hy3 | ✓ Link | 0.28 | Hy3 | 2026-07-02 |
| Ornith-1.5: From Self-Scaffolding to Self-Improvement | | 0.22 | Ornith-1.5-35B-A3B (Claude Code) | 2026-08-20 |
| CodeMidas: Scaling Agentic Coding RL Environments from Code Itself | | 0.217 | MiMo-V2.5 (CodeMidas RL) | 2026-09-18 |
| GLM-5.1: Towards Long-Horizon Tasks | ✓ Link | 0.18 | GLM-5.1 (Max) (MiniSWE-Agent) | 2026-04-03 |
| MindForge: Teaching Small Language Models Whole-Life-Cycle Software Engineering via Source-Free Program Synthesis | | 0.1592 | MindForge-27B (Mini-SWE-Agent) | 2026-07-29 |
| DeepSWE Leaderboard | | 0.1173 | Gemini 3.1 Pro Preview (high) (MiniSWE-Agent) | 2026-02-19 |