MMODELYST
Capabilities/Language & Instruction
CAP

Language & Instruction

League table — best at language & instruction · 30
mean percentile across this capability's benchmarks
#ModelLeagueCoverage$/1Mtok/s
1Llama-3.3-70B-Instruct100.01/1
2Llama-3.1-70B-Instruct99.91/1
3calme-2.1-qwen2.5-72b99.91/1
4Llama-3.1-SauerkrautLM-70b-Instruct99.81/1
5Qwen2.5-72B-Instruct99.71/1
6DeepSeek-V399.71/1
7calme-2.3-llama3.1-70b99.61/1
8calme-2.2-llama3.1-70b99.51/1
9Qwen2.5-72B-Instruct-abliterated99.51/1
10ZYH-LLM-Qwen2.5-14B-V399.41/1
11orca_mini_v8_1_70b99.31/1
12RYS-Llama3.1-Large99.31/1
13Gilgamesh-72B99.21/1
14calme-2.2-qwen2.5-72b99.11/1
15test-2.5-72B99.11/1
16calme-2.1-llama3.1-70b99.01/1
17Rombos-LLM-V2.6-Qwen-14b98.91/1
18Qwen2.5-95B-Instruct98.91/1
19virtuoso-small-v2-tensopolis-v198.81/1
20solar-pro-preview-instruct98.71/1
21Qwen2.5-14B-Instruct-1M-GRPO-Reasoning98.71/1
22Qwen2.5-14B-Instruct-1M98.61/1
23Mistral-Large-Instruct-241198.51/1
24Qwen2.5-14B-1M-YOYO-V398.41/1
25Qwen2.5-14B-YOYO-V498.41/1
26EXAONE-3.5-32B-Instruct98.31/1
27orca_mini_v9_2_70b98.21/1
28Llama-3.1-Tulu-3-70B98.21/1
29ZYH-LLM-Qwen2.5-14B-V498.11/1
30Qwen2.5-32B-Instruct98.01/1

League = a model's mean percentile across the 1 benchmarkin this capability (best score per benchmark; partial coverage shown). Price & speed via Artificial Analysis.

Benchmarks in this capability (1)
IFEval1488 scores