Best local LLMs by hardware tier
Rankings only make sense once you fix the hardware. Pick a tier below — the leaderboard re-sorts to the models the community actually runs there, weighted by report count, fastest observed tokens-per-second, and recency.
AMD discrete, Intel & CPU-only
AMD discrete cards (no Strix Halo), Intel, and CPU-only inference. Ranked from 31 reports.
| # | Model family | Best variant tested | Reports | Fastest t/s |
|---|---|---|---|---|
| 1 | Gemma 4Google DeepMind — | — | 19 | 68.8 |
| 2 | Qwen3.6Alibaba IQ4_XS | IQ4_XS | 11 | 55.0 |
| 3 | Kimi K2.6Moonshot AI — · on AMD MI50 32GB | — | 1 | 9.7 |
How we rank
A single global "best models" list doesn't really exist — what runs well on a 5090 is often unrunnable on a 4060, and a 7B that screams on an M3 Max is usually a poor pick on an H100. So we fix the hardware first, then rank the families that actually have community reports on it. The score blends popularity (log-scaled report count), fastest observed tokens-per-second normalized within the bucket, recency (90-day half-life), and a small bias for rows where we know the variant + quant + GPU cleanly. Click into a family for the full breakdown of records.