# Natural Language Understanding on italic fast Source: [Natural Language Understanding on italic fast]() Collection index: [llms.txt]() Data updated: 2026-05-11T10:11:09.629Z Evaluation of Natural Language Understanding capabilities on the italic fast dataset, typically measured by Exact Match accuracy. 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 | Natural Language Understanding | | Dataset (including any stated split) | italic fast | | Metric | Exact Match (EM) | | 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 - [Exact Match (EM)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Exact Match (EM) | Result date | Report paper | | --- | --- | --- | --- | | GPT-5 nano | 86.1 | Not recorded | 6a0162badbc836b87d82332a | | Ministral-3-8B-Instruct-2512-BF16 | 80.2 | Not recorded | 6a0162badbc836b87d82332a | | gemma-3-12b-it | 76.5 | Not recorded | 6a0162badbc836b87d82332a | | Qwen3-8B | 73 | Not recorded | 6a0162badbc836b87d82332a | | LLaMAntino-3-ANITA-8B-Inst-DPO-ITA | 69.9 | Not recorded | 6a0162badbc836b87d82332a | | Llama-3.1-8B-Instruct | 67.5 | Not recorded | 6a0162badbc836b87d82332a | | Velvet-14B | 67.2 | Not recorded | 6a0162badbc836b87d82332a | | gpt-oss-20b | 67.2 | Not recorded | 6a0162badbc836b87d82332a | | Qwen3-4B | 66.7 | Not recorded | 6a0162badbc836b87d82332a | | gemma-3-4b-it | 62.8 | Not recorded | 6a0162badbc836b87d82332a | | FastwebMIIA-7B | 61.7 | Not recorded | 6a0162badbc836b87d82332a | | EngGPT2-16B-A3B | 60.2 | Not recorded | 6a0162badbc836b87d82332a | | Llama-3.2-3B-Instruct | 57.4 | Not recorded | 6a0162badbc836b87d82332a | | Moonlight-16B-A3B-Instruct | 57 | Not recorded | 6a0162badbc836b87d82332a | | deepseek-moe-16b-chat | 48.5 | Not recorded | 6a0162badbc836b87d82332a | | Minerva-7B-instruct-v1.0 | 34.7 | Not recorded | 6a0162badbc836b87d82332a | ### GPT-5 nano Exact Match (EM): 86.1. Result ID: 6a01ab3c47f34ce924e41ea1. [Reporting paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Ministral-3-8B-Instruct-2512-BF16 Exact Match (EM): 80.2. Result ID: 6a01ab3c47f34ce924e41ea0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### gemma-3-12b-it Exact Match (EM): 76.5. Result ID: 6a01ab3c47f34ce924e41e9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Qwen3-8B Exact Match (EM): 73. Result ID: 6a01ab3c47f34ce924e41e9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### LLaMAntino-3-ANITA-8B-Inst-DPO-ITA Exact Match (EM): 69.9. Result ID: 6a01ab3c47f34ce924e41e99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Llama-3.1-8B-Instruct Exact Match (EM): 67.5. Result ID: 6a01ab3c47f34ce924e41e9d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Velvet-14B Exact Match (EM): 67.2. Result ID: 6a01ab3c47f34ce924e41e98. [Reporting paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### gpt-oss-20b Exact Match (EM): 67.2. Result ID: 6a01ab3c47f34ce924e41ea3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Qwen3-4B Exact Match (EM): 66.7. Result ID: 6a01ab3c47f34ce924e41e9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### gemma-3-4b-it Exact Match (EM): 62.8. Result ID: 6a01ab3c47f34ce924e41e9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### FastwebMIIA-7B Exact Match (EM): 61.7. Result ID: 6a01ab3c47f34ce924e41e96. [Reporting paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### EngGPT2-16B-A3B Exact Match (EM): 60.2. Result ID: 6a01ab3c47f34ce924e41e95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Llama-3.2-3B-Instruct Exact Match (EM): 57.4. Result ID: 6a01ab3c47f34ce924e41e9c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Moonlight-16B-A3B-Instruct Exact Match (EM): 57. Result ID: 6a01ab3c47f34ce924e41ea4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### deepseek-moe-16b-chat Exact Match (EM): 48.5. Result ID: 6a01ab3c47f34ce924e41ea2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. ### Minerva-7B-instruct-v1.0 Exact Match (EM): 34.7. Result ID: 6a01ab3c47f34ce924e41e97. [Reporting paper]() Recorded evaluation configuration: {"integration":"lm\_evaluation\_harness"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.