# Speech-to-Speech Question-Answering on LlamaQ Source: [Speech-to-Speech Question-Answering on LlamaQ]() Collection index: [llms.txt]() Data updated: 2026-07-08T08:02:38.464Z Evaluation of model performance on the LlamaQ dataset for Speech-to-Speech Question-Answering, measured using 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 | Speech-to-Speech Question-Answering | | Dataset (including any stated split) | LlamaQ | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | MinMo | 76.5 | Not recorded | 69802f80a8b01e76cc8e15cc | | DIFFUSPEECH | 72 | Not recorded | 69802f80a8b01e76cc8e15cc | | Llama-Omni2 | 70.3 | Not recorded | 69802f80a8b01e76cc8e15cc | | Moshi | 68.3 | Not recorded | 69802f80a8b01e76cc8e15cc | | Lychee-FD | 65.3 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | Fun-Audio-Chat | 64.3 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | StepAudio-2-mini | 62 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | SALMONN-omni | 61.7 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | Lychee-FD (w/o Sem-Channel) | 61 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | Moshi | 54.7 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | VITA 1.5 | 51 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | Freeze-Omni | 50.7 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | SpiritLM | 40.1 | Not recorded | 69802f80a8b01e76cc8e15cc | | FLM-audio | 36.7 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | Lychee-FD (w/o Param-Sep) | 36 | Not recorded | 6a4dd9dd4da6617fe9198e5a | | SpeechGPT | 21.6 | Not recorded | 69802f80a8b01e76cc8e15cc | | dGSLM | 1.3 | Not recorded | 6a4dd9dd4da6617fe9198e5a | ### MinMo Accuracy: 76.5. Result ID: 69965cb55698665ec4a26158. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"AR","S→S capability":true}. This record does not assert independent reproduction. ### DIFFUSPEECH Accuracy: 72. Result ID: 69965cb55698665ec4a2615a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Diff.","S→S capability":true}. This record does not assert independent reproduction. ### Llama-Omni2 Accuracy: 70.3. Result ID: 69965cb55698665ec4a26159. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"AR","S→S capability":true}. This record does not assert independent reproduction. ### Moshi Accuracy: 68.3. Result ID: 69965cb55698665ec4a26157. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"AR","S→S capability":true}. This record does not assert independent reproduction. ### Lychee-FD Accuracy: 65.3. Result ID: 6a4e040f60b88bb339d7b53d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### Fun-Audio-Chat Accuracy: 64.3. Result ID: 6a4e040f60b88bb339d7b53b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### StepAudio-2-mini Accuracy: 62. Result ID: 6a4e040f60b88bb339d7b53c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Half-duplex"}. This record does not assert independent reproduction. ### SALMONN-omni Accuracy: 61.7. Result ID: 6a4e040f60b88bb339d7b53a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### Lychee-FD (w/o Sem-Channel) Accuracy: 61. Result ID: 6a4e040f60b88bb339d7b53e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### Moshi Accuracy: 54.7. Result ID: 6a4e040f60b88bb339d7b539. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### VITA 1.5 Accuracy: 51. Result ID: 6a4e040f60b88bb339d7b536. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"System-level"}. This record does not assert independent reproduction. ### Freeze-Omni Accuracy: 50.7. Result ID: 6a4e040f60b88bb339d7b535. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"System-level"}. This record does not assert independent reproduction. ### SpiritLM Accuracy: 40.1. Result ID: 69965cb55698665ec4a26156. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"AR","S→S capability":true}. This record does not assert independent reproduction. ### FLM-audio Accuracy: 36.7. Result ID: 6a4e040f60b88bb339d7b538. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### Lychee-FD (w/o Param-Sep) Accuracy: 36. Result ID: 6a4e040f60b88bb339d7b53f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. ### SpeechGPT Accuracy: 21.6. Result ID: 69965cb55698665ec4a26155. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"AR","S→S capability":true}. This record does not assert independent reproduction. ### dGSLM Accuracy: 1.3. Result ID: 6a4e040f60b88bb339d7b537. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Native"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.