OpenCodePapers

code-generation-on-terminal-bench-2-0-terminus-2

Code IntelligenceCode Generation
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAccuracyModelNameReleaseDate
Introducing Claude Opus 4.769.4Claude Opus 4.7 (Terminus-2)2026-04-16
Gemini 3.1 Pro: A smarter model for your most complex tasks68.5Gemini 3.1 Pro (Terminus-2)2026-02-19
Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process Rewards✓ Link68.2LAR-27B (Terminus-2)2026-10-15
Introducing Claude Opus 4.665.4Claude Opus 4.6 (Terminus-2)2026-02-05
Introducing Claude Opus 4.559.3Claude Opus 4.5 (Terminus-2)2025-11-24
Qwen3.6✓ Link59.3Qwen3.6-27B (Terminus-2)2026-04-21
The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence✓ Link57.0MiniMax-M2.7 (Terminus-2)2026-05-26
A new era of intelligence with Gemini 354.2Gemini 3 Pro (Terminus-2)2025-11-18
Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments52.8Terminal-Universe-27B (Terminus 2)2026-09-03
Qwen3.5: Towards Native Multimodal Agents✓ Link52.5Qwen3.5-397B-A17B (Terminus-2)2026-02-16
Qwen3.6✓ Link51.5Qwen3.6-35B-A3B (Terminus-2)2026-04-21
Kimi K2.5: Visual Agentic Intelligence✓ Link50.8Kimi K2.5 (Terminus-2)2026-02-02
Introducing Claude Sonnet 4.550.0Claude Sonnet 4.5 (Terminus-2)2025-09-29
Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model✓ Link44.1Nanbeige4.2-3B (Terminus-2)2026-07-21
Qwen3-Coder-Next Technical Report✓ Link36.2Qwen3-Coder-Next (Terminus2-json)2026-02-28
Terminal-Bench Leaderboard✓ Link33.4GLM-4.7 (Terminus-2)2025-08-08
CLI-Universe: Towards Verifiable Task Synthesis Engine for Terminal Agents33.4CLI-Universe-32B (Terminus 2)2026-06-22
On Data Engineering for Scaling LLM Terminal Capabilities27.4Nemotron-Terminal-32B (Terminus-2)2026-02-24
OpenThoughts-Agent: Data Recipes for Agentic Models✓ Link26.2OpenThinkerAgent-32B (Terminus-2)2026-06-23
Terminal-Bench Leaderboard✓ Link24.5GLM-4.6 (Terminus-2)2025-08-08
Unified Audio Intelligence Without Regressing on Text Intelligence19.1Audex 30B-A3B (Terminus-2)2026-07-06