# Reaction Feasibility Prediction on FREA Infeasible 1.0 (test) Source: [Reaction Feasibility Prediction on FREA Infeasible 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-11T08:16:19.884Z Evaluation of models performing Reaction Feasibility Prediction on the test split of the FREA Infeasible 1.0 dataset. 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 | Reaction Feasibility Prediction | | Dataset (including any stated split) | FREA Infeasible 1.0 (test) | | Metric | Accuracy (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 - [Accuracy (%)](): Higher is better; unit: PERCENT. - [F1 Score (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (%) | Result date | Report paper | | --- | --- | --- | --- | | Theoretical Upper Bound | 100 | Not recorded | 6a016299dbc836b87d823180 | | Claude-Sonnet-4.6 | 96.53 | Not recorded | 6a016299dbc836b87d823180 | | Chemformer | 95.53 | Not recorded | 6a016299dbc836b87d823180 | | BERT | 95.5 | Not recorded | 6a016299dbc836b87d823180 | | DRFP | 95.1 | Not recorded | 6a016299dbc836b87d823180 | | ToolEyes | 94.77 | Not recorded | 6a016299dbc836b87d823180 | | Molecular | 93.27 | Not recorded | 6a016299dbc836b87d823180 | | HarderMoE | 92.97 | Not recorded | 6a016299dbc836b87d823180 | | ARMOR | 91.67 | Not recorded | 6a016299dbc836b87d823180 | | Weighted Voting | 91.2 | Not recorded | 6a016299dbc836b87d823180 | | PromptingLlama | 84.67 | Not recorded | 6a016299dbc836b87d823180 | | DeepSeek-v4-flash | 84.67 | Not recorded | 6a016299dbc836b87d823180 | | KNORA-U | 83.37 | Not recorded | 6a016299dbc836b87d823180 | | Majority Voting | 83.23 | Not recorded | 6a016299dbc836b87d823180 | | KNORA-E | 83.23 | Not recorded | 6a016299dbc836b87d823180 | | Dora\_xgb | 79.6 | Not recorded | 6a016299dbc836b87d823180 | | DES-KNN | 75.03 | Not recorded | 6a016299dbc836b87d823180 | | StaticSel | 66.37 | Not recorded | 6a016299dbc836b87d823180 | | Random Selection | 66.29 | Not recorded | 6a016299dbc836b87d823180 | | MolecularT3Q | 64.1 | Not recorded | 6a016299dbc836b87d823180 | | GPT-5.4-mini | 64.03 | Not recorded | 6a016299dbc836b87d823180 | | PromptingT3Q | 61.83 | Not recorded | 6a016299dbc836b87d823180 | | ChemformerT3Q | 59.63 | Not recorded | 6a016299dbc836b87d823180 | | DES-Clustering | 59.5 | Not recorded | 6a016299dbc836b87d823180 | | ICLT3Q | 50.5 | Not recorded | 6a016299dbc836b87d823180 | | MolecularLlama | 36.9 | Not recorded | 6a016299dbc836b87d823180 | | ChemformerLlama | 30.4 | Not recorded | 6a016299dbc836b87d823180 | | ICLLlama | 20.3 | Not recorded | 6a016299dbc836b87d823180 | ### Theoretical Upper Bound Accuracy (%): 100. Result ID: 6a01903f47f34ce924e34a8a. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Claude-Sonnet-4.6 Accuracy (%): 96.53. Result ID: 6a01903f47f34ce924e34a97. [Reporting paper]() Recorded evaluation configuration: {"Category":"LLM-based Methods"}. This record does not assert independent reproduction. ### Chemformer Accuracy (%): 95.53. Result ID: 6a01903f47f34ce924e34a81. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods"}. This record does not assert independent reproduction. ### BERT Accuracy (%): 95.5. Result ID: 6a01903f47f34ce924e34a7d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Single-Tool Methods"}. This record does not assert independent reproduction. ### DRFP Accuracy (%): 95.1. Result ID: 6a01903f47f34ce924e34a7e. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods"}. This record does not assert independent reproduction. ### ToolEyes Accuracy (%): 94.77. Result ID: 6a01903f47f34ce924e34a94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"LLM-based Methods"}. This record does not assert independent reproduction. ### Molecular Accuracy (%): 93.27. Result ID: 6a01903f47f34ce924e34a80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods"}. This record does not assert independent reproduction. ### HarderMoE Accuracy (%): 92.97. Result ID: 6a01903f47f34ce924e34a93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Dynamic Methods"}. This record does not assert independent reproduction. ### ARMOR Accuracy (%): 91.67. Result ID: 6a01903f47f34ce924e34a98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"LLM-based Methods","Backbone":"T3Q-Qwen-14B"}. This record does not assert independent reproduction. ### Weighted Voting Accuracy (%): 91.2. Result ID: 6a01903f47f34ce924e34a8d. [Reporting paper]() Recorded evaluation configuration: {"Category":"Statistical Methods"}. This record does not assert independent reproduction. ### PromptingLlama Accuracy (%): 84.67. Result ID: 6a01903f47f34ce924e34a83. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"Llama"}. This record does not assert independent reproduction. ### DeepSeek-v4-flash Accuracy (%): 84.67. Result ID: 6a01903f47f34ce924e34a96. [Reporting paper]() Recorded evaluation configuration: {"Category":"LLM-based Methods"}. This record does not assert independent reproduction. ### KNORA-U Accuracy (%): 83.37. Result ID: 6a01903f47f34ce924e34a92. [Reporting paper]() Recorded evaluation configuration: {"Category":"Dynamic Methods"}. This record does not assert independent reproduction. ### Majority Voting Accuracy (%): 83.23. Result ID: 6a01903f47f34ce924e34a8c. [Reporting paper]() Recorded evaluation configuration: {"Category":"Statistical Methods"}. This record does not assert independent reproduction. ### KNORA-E Accuracy (%): 83.23. Result ID: 6a01903f47f34ce924e34a91. [Reporting paper]() Recorded evaluation configuration: {"Category":"Dynamic Methods"}. This record does not assert independent reproduction. ### Dora\_xgb Accuracy (%): 79.6. Result ID: 6a01903f47f34ce924e34a7f. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods"}. This record does not assert independent reproduction. ### DES-KNN Accuracy (%): 75.03. Result ID: 6a01903f47f34ce924e34a8f. [Reporting paper]() Recorded evaluation configuration: {"Category":"Dynamic Methods"}. This record does not assert independent reproduction. ### StaticSel Accuracy (%): 66.37. Result ID: 6a01903f47f34ce924e34a8b. [Reporting paper]() Recorded evaluation configuration: {"Category":"Statistical Methods"}. This record does not assert independent reproduction. ### Random Selection Accuracy (%): 66.29. Result ID: 6a01903f47f34ce924e34a8e. [Reporting paper]() Recorded evaluation configuration: {"Category":"Statistical Methods"}. This record does not assert independent reproduction. ### MolecularT3Q Accuracy (%): 64.1. Result ID: 6a01903f47f34ce924e34a86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"T3Q"}. This record does not assert independent reproduction. ### GPT-5.4-mini Accuracy (%): 64.03. Result ID: 6a01903f47f34ce924e34a95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"LLM-based Methods"}. This record does not assert independent reproduction. ### PromptingT3Q Accuracy (%): 61.83. Result ID: 6a01903f47f34ce924e34a82. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"T3Q"}. This record does not assert independent reproduction. ### ChemformerT3Q Accuracy (%): 59.63. Result ID: 6a01903f47f34ce924e34a88. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"T3Q"}. This record does not assert independent reproduction. ### DES-Clustering Accuracy (%): 59.5. Result ID: 6a01903f47f34ce924e34a90. [Reporting paper]() Recorded evaluation configuration: {"Category":"Dynamic Methods"}. This record does not assert independent reproduction. ### ICLT3Q Accuracy (%): 50.5. Result ID: 6a01903f47f34ce924e34a84. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"T3Q"}. This record does not assert independent reproduction. ### MolecularLlama Accuracy (%): 36.9. Result ID: 6a01903f47f34ce924e34a87. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"Llama"}. This record does not assert independent reproduction. ### ChemformerLlama Accuracy (%): 30.4. Result ID: 6a01903f47f34ce924e34a89. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"Llama"}. This record does not assert independent reproduction. ### ICLLlama Accuracy (%): 20.3. Result ID: 6a01903f47f34ce924e34a85. [Reporting paper]() Recorded evaluation configuration: {"Category":"Single-Tool Methods","Backbone":"Llama"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.