# Retrosynthesis on Chemical Reaction-related Tasks (test) Source: [Retrosynthesis on Chemical Reaction-related Tasks (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:42:26.067Z Evaluation of models performing retrosynthesis prediction on the test split of the Chemical Reaction-related Tasks dataset, measured using metrics including exact match, string similarity (BLEU, Levenshtein), and chemical fingerprint similarity (RDK, MACCS, Morgan FTS). 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 | Retrosynthesis | | Dataset (including any stated split) | Chemical Reaction-related Tasks (test) | | Metric | Exact Match | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [Exact Match](): Higher is better; unit: PERCENT. - [BLEU](): Higher is better; unit: PERCENT. - [Levenshtein Distance](): Lower is better; unit: SCALAR. - [RDK FTS](): Higher is better; unit: PERCENT. - [MACCS FTS](): Higher is better; unit: PERCENT. - [MORGAN FTS](): Higher is better; unit: PERCENT. - [Validity](): Higher is better; unit: PERCENT. ## Recorded results | Method | Exact Match | Result date | Report paper | | --- | --- | --- | --- | | BioT5+ | 64.2 | Not recorded | 68fa5f5b1e209131ce9e9311 | | InstructMol-GS-6.9B | 40.7 | Not recorded | 68fa5f5b1e209131ce9e9311 | | Text+Chem T5-223M | 14.1 | Not recorded | 68fa5f5b1e209131ce9e9311 | | InstructMol-G-6.9B | 11.4 | Not recorded | 68fa5f5b1e209131ce9e9311 | | Mol-Instructions-7B | 0.9 | Not recorded | 68fa5f5b1e209131ce9e9311 | | Llama-7B | 0 | Not recorded | 68fa5f5b1e209131ce9e9311 | | Galactica-6.7B | 0 | Not recorded | 68fa5f5b1e209131ce9e9311 | | Llama-7B\*(LORA) | 0 | Not recorded | 68fa5f5b1e209131ce9e9311 | ### BioT5+ Exact Match: 64.2. Result ID: 698bc4817a967eddb771355b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### InstructMol-GS-6.9B Exact Match: 40.7. Result ID: 698bc4817a967eddb771355a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Vicuna","Parameter count":"6.9B"}. This record does not assert independent reproduction. ### Text+Chem T5-223M Exact Match: 14.1. Result ID: 698bc4817a967eddb7713556. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"T5","Parameter count":"223M"}. This record does not assert independent reproduction. ### InstructMol-G-6.9B Exact Match: 11.4. Result ID: 698bc4817a967eddb7713559. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Vicuna","Parameter count":"6.9B"}. This record does not assert independent reproduction. ### Mol-Instructions-7B Exact Match: 0.9. Result ID: 698bc4817a967eddb7713557. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-7B","Parameter count":"7B","Multi-task instruction tuning":true}. This record does not assert independent reproduction. ### Llama-7B Exact Match: 0. Result ID: 698bc4817a967eddb7713554. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama","Parameter count":"7B","Zero-shot":true}. This record does not assert independent reproduction. ### Galactica-6.7B Exact Match: 0. Result ID: 698bc4817a967eddb7713555. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Galactica","Parameter count":"6.7B","Zero-shot":true}. This record does not assert independent reproduction. ### Llama-7B\*(LORA) Exact Match: 0. Result ID: 698bc4817a967eddb7713558. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama","Parameter count":"7B","LoRA tuning":true}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.