Leaderboard
On-device LLM performance rankings powered by Glicko-2
OnePlus 11 5G
AndroidRank
#96
Rating
1,644
±16 RD
Win Rate
63.9%
Conservative Rating
1,612
TG Rating
1,658
PP Rating
1,574
Matches
1,040
Record
665W – 375L
Models Tested
| Model | TG Median (tok/s) | PP Median (tok/s) | TG Best | PP Best | Runs |
|---|---|---|---|---|---|
| qwen2.5-1.5b-instruct-q8_0 | 17.22 | 85.24 | 17.22 | 85.24 | 1 |
| llama-3.2-1b-instruct-q8_0 | 15.20 | 68.79 | 15.20 | 68.79 | 1 |
| DeepSeek-R1-Distill-Qwen-1.5B-Q4_K_M | 12.32 | 42.73 | 12.32 | 42.73 | 1 |
| gemma-2-2b-it-Q6_K | 9.19 | 19.76 | 9.46 | 21.44 | 5 |
| Gemmasutra-Mini-2B-v1-Q6_K | 8.18 | 15.84 | 8.18 | 15.84 | 1 |
| Llama-3.2-3B-Instruct-uncensored-Q8_0 | 7.76 | 29.49 | 7.76 | 29.49 | 1 |
| Phi-4-mini-reasoning-Q4_1 | 7.65 | 20.07 | 7.65 | 20.07 | 1 |
| Llama-3.2-1B-Instruct-F16 | 7.39 | 13.48 | 7.39 | 13.48 | 1 |
| Phi-3.5-mini-instruct.Q4_K_M | 6.87 | 13.25 | 7.40 | 13.55 | 2 |
| Qwen2.5-7B-Instruct.Q4_K_M | 4.48 | 8.56 | 4.48 | 8.56 | 1 |
| DeepSeek-R1-Distill-Llama-8B-Q3_K_M | 3.77 | 6.00 | 3.77 | 6.00 | 1 |
| mistral-7b-uncensored-Q3_K_M | 3.62 | 6.30 | 3.62 | 6.30 | 1 |
| qwen2.5-3b-instruct-q5_k_m | 3.27 | 10.42 | 5.39 | 10.91 | 2 |
| DeepSeek-R1-Distill-Qwen-14B-IQ2_XS | 1.23 | 1.69 | 1.23 | 1.69 | 1 |
Head-to-Head Record
1–50 of 271 rows
1 / 6
Performance by App Version
ImprovedRegressed