# Web Task Completion on MiniWoB++ Held-out Tasks (test) Source: [Web Task Completion on MiniWoB++ Held-out Tasks (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:05:27.433Z Evaluation of web agent performance on the specific held-out test tasks within the MiniWoB++ environment, measured by success rate. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Web Task Completion | | Dataset (including any stated split) | MiniWoB++ Held-out Tasks (test) | | Metric | Success Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Success Rate | Result date | Report paper | | --- | --- | --- | --- | | Claude-3.5-Sonnet | 70.4 | Not recorded | 68fa79d41e209131cea7f95b | | GPT-4o-Mini | 66.1 | Not recorded | 68fa79d41e209131cea7f95b | | o1-Mini | 66.1 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-70B-Instruct | 65.2 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-405B-Instruct | 65.2 | Not recorded | 68fa79d41e209131cea7f95b | | GPT-4o | 64.3 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-8B SFT+RL (Ours) | 63.3 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.3-70B Instruct (Teacher) | 61.9 | Not recorded | 68fa79d41e209131cea7f95b | | Qwen-2.5-7B SFT+RL (Ours) | 61.5 | Not recorded | 68fa79d41e209131cea7f95b | | Qwen-2.5-72B Instruct (Teacher) | 59 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-8B SFT (Ours) | 56.7 | Not recorded | 68fa79d41e209131cea7f95b | | Qwen-2.5-7B SFT (Ours) | 56.5 | Not recorded | 68fa79d41e209131cea7f95b | | Qwen-2.5-7B RL (Ours) | 53 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-8B RL (Ours) | 43.5 | Not recorded | 68fa79d41e209131cea7f95b | | Qwen-2.5-7B Instruct (Student) | 37 | Not recorded | 68fa79d41e209131cea7f95b | | Llama-3.1-8B Instruct (Student) | 36.4 | Not recorded | 68fa79d41e209131cea7f95b | ### Claude-3.5-Sonnet Success Rate: 70.4. Result ID: 699a631c5698665ec4eb31b8. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Claude-3.5-Sonnet","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### GPT-4o-Mini Success Rate: 66.1. Result ID: 699a631c5698665ec4eb31ba. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"GPT-4o-Mini","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### o1-Mini Success Rate: 66.1. Result ID: 699a631c5698665ec4eb31bc. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"o1-Mini","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### Llama-3.1-70B-Instruct Success Rate: 65.2. Result ID: 699a631c5698665ec4eb31bb. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-70B","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### Llama-3.1-405B-Instruct Success Rate: 65.2. Result ID: 699a631c5698665ec4eb31bd. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-405B","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### GPT-4o Success Rate: 64.3. Result ID: 699a631c5698665ec4eb31b9. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"GPT-4o","Training Protocol":"Instruct"}. This record does not assert independent reproduction. ### Llama-3.1-8B SFT+RL (Ours) Success Rate: 63.3. Result ID: 699a631c5698665ec4eb31c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-8B","Training Protocol":"SFT+RL"}. This record does not assert independent reproduction. ### Llama-3.3-70B Instruct (Teacher) Success Rate: 61.9. Result ID: 699a631c5698665ec4eb31be. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.3-70B","Training Protocol":"Instruct","Model Role":"Teacher"}. This record does not assert independent reproduction. ### Qwen-2.5-7B SFT+RL (Ours) Success Rate: 61.5. Result ID: 699a631c5698665ec4eb31c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen-2.5-7B","Training Protocol":"SFT+RL"}. This record does not assert independent reproduction. ### Qwen-2.5-72B Instruct (Teacher) Success Rate: 59. Result ID: 699a631c5698665ec4eb31c3. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen-2.5-72B","Training Protocol":"Instruct","Model Role":"Teacher"}. This record does not assert independent reproduction. ### Llama-3.1-8B SFT (Ours) Success Rate: 56.7. Result ID: 699a631c5698665ec4eb31c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-8B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### Qwen-2.5-7B SFT (Ours) Success Rate: 56.5. Result ID: 699a631c5698665ec4eb31c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen-2.5-7B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### Qwen-2.5-7B RL (Ours) Success Rate: 53. Result ID: 699a631c5698665ec4eb31c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen-2.5-7B","Training Protocol":"RL"}. This record does not assert independent reproduction. ### Llama-3.1-8B RL (Ours) Success Rate: 43.5. Result ID: 699a631c5698665ec4eb31c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-8B","Training Protocol":"RL"}. This record does not assert independent reproduction. ### Qwen-2.5-7B Instruct (Student) Success Rate: 37. Result ID: 699a631c5698665ec4eb31c4. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Qwen-2.5-7B","Training Protocol":"Instruct","Model Role":"Student"}. This record does not assert independent reproduction. ### Llama-3.1-8B Instruct (Student) Success Rate: 36.4. Result ID: 699a631c5698665ec4eb31bf. [Reporting paper]() Recorded evaluation configuration: {"Backbone Model":"Llama-3.1-8B","Training Protocol":"Instruct","Model Role":"Student"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.