# General LLM Evaluation on Instruction-Following, Mathematics, and Commonsense Reasoning Combined Source: [General LLM Evaluation on Instruction-Following, Mathematics, and Commonsense Reasoning Combined]() Collection index: [llms.txt]() Data updated: 2026-03-12T08:13:49.473Z A comprehensive evaluation suite measuring general LLM performance across instruction following, mathematical problem-solving, and commonsense reasoning capabilities, summarized by an average score. 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 | General LLM Evaluation | | Dataset (including any stated split) | Instruction-Following, Mathematics, and Commonsense Reasoning Combined | | Metric | Average Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Average Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen2.5 7B-PC | 57 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 3.1 8B-PC | 56 | Not recorded | 68fa78b81e209131cea7a080 | | Olmo 3 7B-PC | 55 | Not recorded | 68fa78b81e209131cea7a080 | | Llama 3.1 8B-PC | 54 | Not recorded | 68fa78b81e209131cea7a080 | | Qwen2.5 7B-NL | 53 | Not recorded | 68fa78b81e209131cea7a080 | | Olmo 3 7B-NL | 50 | Not recorded | 68fa78b81e209131cea7a080 | | Mistral 7B v0.3-PC | 48 | Not recorded | 68fa78b81e209131cea7a080 | | Qwen2.5 7B-CP | 47 | Not recorded | 68fa78b81e209131cea7a080 | | Llama 3.1 8B-NL | 45 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 8B Code-PC | 43 | Not recorded | 68fa78b81e209131cea7a080 | | Olmo 3 7B-CP | 39 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 3.1 8B-CP | 37 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 8B Code-NL | 36 | Not recorded | 68fa78b81e209131cea7a080 | | Mistral 7B v0.3-NL | 35 | Not recorded | 68fa78b81e209131cea7a080 | | Llama 3.1 8B-CP | 34 | Not recorded | 68fa78b81e209131cea7a080 | | Mistral 7B v0.3-CP | 33 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 8B Code-CP | 29 | Not recorded | 68fa78b81e209131cea7a080 | | Granite 3.1 8B-NL | 26 | Not recorded | 68fa78b81e209131cea7a080 | ### Qwen2.5 7B-PC Average Score: 57. Result ID: 69b274836dbb7075b95cebf6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5 7B","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 3.1 8B-PC Average Score: 56. Result ID: 69b274836dbb7075b95cebfc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Granite 3.1 8B","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Olmo 3 7B-PC Average Score: 55. Result ID: 69b274836dbb7075b95cebff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Olmo 3 7B","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Llama 3.1 8B-PC Average Score: 54. Result ID: 69b274836dbb7075b95cebf0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Qwen2.5 7B-NL Average Score: 53. Result ID: 69b274836dbb7075b95cebf5. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5 7B","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Olmo 3 7B-NL Average Score: 50. Result ID: 69b274836dbb7075b95cebfe. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Olmo 3 7B","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Mistral 7B v0.3-PC Average Score: 48. Result ID: 69b274836dbb7075b95cebf3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral 7B v0.3","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Qwen2.5 7B-CP Average Score: 47. Result ID: 69b274836dbb7075b95cebf4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5 7B","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Llama 3.1 8B-NL Average Score: 45. Result ID: 69b274836dbb7075b95cebef. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 8B Code-PC Average Score: 43. Result ID: 69b274836dbb7075b95cebf9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Granite 8B Code","Strategy":"Pseudo-code tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Olmo 3 7B-CP Average Score: 39. Result ID: 69b274836dbb7075b95cebfd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Olmo 3 7B","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 3.1 8B-CP Average Score: 37. Result ID: 69b274836dbb7075b95cebfa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Granite 3.1 8B","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 8B Code-NL Average Score: 36. Result ID: 69b274836dbb7075b95cebf8. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Granite 8B Code","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Mistral 7B v0.3-NL Average Score: 35. Result ID: 69b274836dbb7075b95cebf2. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Mistral 7B v0.3","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Llama 3.1 8B-CP Average Score: 34. Result ID: 69b274836dbb7075b95cebee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama 3.1 8B","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Mistral 7B v0.3-CP Average Score: 33. Result ID: 69b274836dbb7075b95cebf1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral 7B v0.3","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 8B Code-CP Average Score: 29. Result ID: 69b274836dbb7075b95cebf7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Granite 8B Code","Strategy":"Code-Prompted","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. ### Granite 3.1 8B-NL Average Score: 26. Result ID: 69b274836dbb7075b95cebfb. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Granite 3.1 8B","Strategy":"NL-tuned","Zero-shot":"true","Greedy decoding":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.