Which local LLMs can the Apple M6 (Mac mini) run?
Unified memory — by default the GPU can use about 70% of it
Assumes 8k context and an f16 KV cache. Verdicts are shown for each memory size.
Specs
- Memory options
- 16 GB · 24 GB · 32 GB
- Memory bandwidth
- 153.6–170.6 GB/s
- FP16 compute
- 5.3 TFLOPS
- Launch year
- 2026
- Macs with this chip
- Mac mini: 16 GB · 24 GB · 32 GB
- Also written as
- M6
Verdicts at a glance
At Q4_K_M (or the closest available quant) with 8k context.
- 16 GBRuns great13 modelsRuns well10 modelsRuns slowly5 modelsWon't run30 models5 more models run at a lower quant.
- 24 GBRuns great17 modelsRuns well13 modelsRuns slowly10 modelsWon't run18 models7 more models run at a lower quant.
- 32 GBRuns great20 modelsRuns well13 modelsRuns slowly15 modelsWon't run10 models1 more model runs at a lower quant.
How to read the verdicts
- Runs great
- Fully on the GPU at 20 tok/s or more
- Runs well
- Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
- Runs slowly
- 2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
- Won't run
- Does not fit, or under 2 tok/s
Every model on the Apple M6
Scroll sideways to see every column.
| Model | Verdict | Speed | Quant | Memory | Runs as | Context | Notes |
|---|---|---|---|---|---|---|---|
| MiniCPM5 1B | Runs great | est. 145.5 tok/s116.4–174.6calibrated estimate ±20% | Q4_K_M | 1.5 / 22.4 GB | Unified memory | up to 128k |
|
| Ling-3.0 Tiny | Runs great | est. 124.4 tok/s87.1–161.7theoretical estimate ±30% | Q4_K_M | 5.5 / 22.4 GB | Unified memory | up to 128k |
|
| Llama 3.2 1B | Runs great | est. 120.2 tok/s96.2–144.3calibrated estimate ±20% | Q4_K_M | 1.7 / 22.4 GB | Unified memory | up to 128k | — |
| EXAONE 4.0 1.2B | Runs great | est. 98.7 tok/s79.0–118.5calibrated estimate ±20% | Q4_K_M | 1.9 / 22.4 GB | Unified memory | up to 32k |
|
| Gemma 4 E2B | Runs great | est. 73.0 tok/s51.1–94.8theoretical estimate ±30% | Q4_K_M | 3.7 / 22.4 GB | Unified memory | up to 128k |
|
| LFM2.5 2.6B | Runs great | est. 71.9 tok/s50.3–93.4theoretical estimate ±30% | Q4_K_M | 2.4 / 22.4 GB | Unified memory | up to 128k |
|
| HyperCLOVA X SEED 1.5B | Runs great | est. 71.4 tok/s57.1–85.7calibrated estimate ±20% | Q4_K_M | 2.4 / 22.4 GB | Unified memory | up to 16k | — |
| Mellum2.1 12B-A2.5B Thinking | Runs great | est. 69.4 tok/s55.6–83.3calibrated estimate ±20% | Q4_K_M | 8.8 / 22.4 GB | Unified memory | up to 128k |
|
| MiniCPM5 2B | Runs great | est. 67.8 tok/s54.2–81.4calibrated estimate ±20% | Q4_K_M | 2.5 / 22.4 GB | Unified memory | up to 32k |
|
| MiniCPM-V 4.7 35B-A3B | Runs great | est. 50.8 tok/s40.6–60.9calibrated estimate ±20% | Q4_K_M | 21.9 / 22.4 GB | Unified memory | up to 16k |
|
| gpt-oss-20b | Runs great | est. 44.8 tok/s35.8–53.7calibrated estimate ±20% | MXFP4 | 12.9 / 22.4 GB | Unified memory | up to 64k |
|
| Spark-X2.5 4B | Runs great | est. 44.7 tok/s35.7–53.6calibrated estimate ±20% | Q4_K_M | 3.5 / 22.4 GB | Unified memory | up to 32k |
|
| Llama 3.2 3B | Runs great | est. 43.8 tok/s35.0–52.6calibrated estimate ±20% | Q4_K_M | 3.5 / 22.4 GB | Unified memory | up to 32k | — |
| Qwen3.5 4B | Runs great | est. 41.8 tok/s33.5–50.2calibrated estimate ±20% | Q4_K_M | 3.7 / 22.4 GB | Unified memory | up to 32k |
|
| Qwen3 30B-A3B (2507) | Runs great | est. 40.4 tok/s32.3–48.5calibrated estimate ±20% | Q4_K_M | 19.9 / 22.4 GB | Unified memory | up to 16k | — |
| Kanana 1.5 15.7B-A3B | Runs great | est. 40.2 tok/s32.1–48.2calibrated estimate ±20% | Q4_K_M | 12.1 / 22.4 GB | Unified memory | up to 32k | — |
| Gemma 4 E4B | Runs great | est. 38.1 tok/s26.7–49.6theoretical estimate ±30% | Q4_K_M | 5.7 / 22.4 GB | Unified memory | up to 32k |
|
| Gemma 4 26B-A4B | Runs great | est. 32.8 tok/s26.3–39.4calibrated estimate ±20% | Q4_K_M | 17.7 / 22.4 GB | Unified memory | up to 16k |
|
| Kanana 1.5 8B | Runs great | est. 21.6 tok/s17.3–26.0calibrated estimate ±20% | Q4_K_M | 6.6 / 22.4 GB | Unified memory | up to 8k | — |
| Llama 3.1 8B | Runs great | est. 21.6 tok/s17.3–26.0calibrated estimate ±20% | Q4_K_M | 6.6 / 22.4 GB | Unified memory | up to 8k | — |
| DeepSeek R1 Distill Llama 8B | Runs well | est. 21.6 tok/s17.3–26.0calibrated estimate ±20% | Q4_K_M | 6.6 / 22.4 GB | Unified memory | up to 32k |
|
| Nanbeige4.2 3B | Runs well | est. 21.2 tok/s14.8–27.5theoretical estimate ±30% | Q4_K_M | 4.7 / 22.4 GB | Unified memory | up to 32k |
|
| Qwen3.5 9B | Runs well | est. 21.1 tok/s16.9–25.3calibrated estimate ±20% | Q4_K_M | 6.7 / 22.4 GB | Unified memory | up to 128k |
|
| DeepSeek R1 0528 Qwen3 8B | Runs well | est. 20.8 tok/s16.6–24.9calibrated estimate ±20% | Q4_K_M | 6.8 / 22.4 GB | Unified memory | up to 32k |
|
| Qwen3 8B | Runs well | est. 20.8 tok/s16.6–24.9calibrated estimate ±20% | Q4_K_M | 6.8 / 22.4 GB | Unified memory | up to 32k |
|
| K2-Horizon 7B | Runs well | est. 19.1 tok/s15.3–22.9calibrated estimate ±20% | Q4_K_M | 7.4 / 22.4 GB | Unified memory | up to 32k |
|
| Gemma 4 12B | Runs well | est. 17.9 tok/s14.3–21.4calibrated estimate ±20% | Q4_K_M | 7.8 / 22.4 GB | Unified memory | up to 128k |
|
| Gemma 3 12B | Runs well | est. 16.5 tok/s13.2–19.9calibrated estimate ±20% | Q4_K_M | 8.4 / 22.4 GB | Unified memory | up to 128k |
|
| HyperCLOVA X SEED Think 14B | Runs well | est. 12.7 tok/s8.9–16.5theoretical estimate ±30% | Q4_K_M | 10.8 / 22.4 GB | Unified memory | up to 8k |
|
| Qwen3 14B | Runs well | est. 12.5 tok/s10.0–15.0calibrated estimate ±20% | Q4_K_M | 10.9 / 22.4 GB | Unified memory | up to 8k |
|
| DeepSeek R1 Distill Qwen 14B | Runs well | est. 12.2 tok/s9.8–14.7calibrated estimate ±20% | Q4_K_M | 11.2 / 22.4 GB | Unified memory | up to 8k |
|
| Phi-4 | Runs well | est. 12.1 tok/s9.7–14.5calibrated estimate ±20% | Q4_K_M | 11.3 / 22.4 GB | Unified memory | up to 16k | — |
| Mistral Small 3.2 24B | Runs well | est. 8.3 tok/s6.6–9.9calibrated estimate ±20% | Q4_K_M | 16.2 / 22.4 GB | Unified memory | up to 8k | — |
| Qwen3.6 35B-A3B | Runs slowly | est. 36.2 tok/s25.3–47.0theoretical estimate ±30% | Q4_K_M | 22.5 GB RAM | CPU only | up to 256k |
|
| Qwen3.5 35B-A3B | Runs slowly | est. 35.7 tok/s25.0–46.4theoretical estimate ±30% | Q4_K_M | 22.8 GB RAM | CPU only | up to 256k |
|
| Nemotron 3.5 Lightning 30B-A3B | Runs slowly | est. 29.7 tok/s20.8–38.6theoretical estimate ±30% | Q4_K_M | 25.5 GB RAM | CPU only | up to 256k |
|
| K2-Horizon MoVA 36B-A4B | Runs slowly | est. 18.3 tok/s12.8–23.8theoretical estimate ±30% | Q4_K_M | 24.0 GB RAM | CPU only | up to 16k |
|
| Gemma 3 27B | Runs slowly | est. 7.5 tok/s6.0–9.0calibrated estimate ±20% | Q4_K_M | 17.8 / 22.4 GB | Unified memory | up to 32k |
|
| Qwen3.6 27B | Runs slowly | est. 7.5 tok/s6.0–9.0calibrated estimate ±20% | Q4_K_M | 17.9 / 22.4 GB | Unified memory | up to 64k |
|
| Muse Glimmer 30B | Runs slowly | est. 7.4 tok/s6.0–8.9calibrated estimate ±20% | Q4_K_M | 18.0 / 22.4 GB | Unified memory | up to 128k |
|
| Qwen3.5 27B | Runs slowly | est. 7.3 tok/s5.9–8.8calibrated estimate ±20% | Q4_K_M | 18.2 / 22.4 GB | Unified memory | up to 32k |
|
| Qwen3.8 27B | Runs slowly | est. 7.2 tok/s5.8–8.7calibrated estimate ±20% | Q4_K_M | 18.6 / 22.4 GB | Unified memory | up to 32k |
|
| Gemma 4 31B | Runs slowly | est. 6.8 tok/s5.5–8.2calibrated estimate ±20% | Q4_K_M | 19.6 / 22.4 GB | Unified memory | up to 32k |
|
| Granite 4.2 30B | Runs slowly | est. 6.5 tok/s5.2–7.8calibrated estimate ±20% | Q4_K_M | 20.4 / 22.4 GB | Unified memory | up to 8k |
|
| EXAONE 4.0 32B | Runs slowly | est. 6.5 tok/s5.2–7.8calibrated estimate ±20% | Q4_K_M | 20.5 / 22.4 GB | Unified memory | up to 32k |
|
| EXAONE 4.5 33B | Runs slowly | est. 6.3 tok/s5.0–7.6calibrated estimate ±20% | Q4_K_M | 21.2 / 22.4 GB | Unified memory | up to 16k |
|
| Qwen3 32B | Runs slowly | est. 3.9 tok/s2.7–5.1theoretical estimate ±30% | Q4_K_M | 21.9 GB RAM | CPU only | up to 16k |
|
| DeepSeek R1 Distill Qwen 32B | Runs slowly | est. 3.9 tok/s2.7–5.0theoretical estimate ±30% | Q4_K_M | 22.0 GB RAM | CPU only | up to 16k |
|
| Llama 3.3 70B | Won't runTry IQ2_M (heavy quality loss): Runs slowly | est. 3.2 tok/s2.2–4.1theoretical estimate ±30% | IQ2_M | 26.8 GB RAM | CPU only | up to 8k | — |
| DeepSeek V4 Flash 0731 | Won't run | — | MXFP4 | needs 155.6 GB | — | — |
|
| GLM-5.3 Flash | Won't run | — | Q4_K_M | needs 182.1 GB | — | — |
|
| gpt-oss-120b | Won't run | — | MXFP4 | needs 64.3 GB | — | — |
|
| Kolibri 1 | Won't run | — | Q4_K_M | needs 50.5 GB | — | — |
|
| Laguna S 2.1 | Won't run | — | Q4_K_M | needs 68.6 GB | — | — |
|
| Ling-3.0 Flash | Won't run | — | Q4_K_M | needs 77.7 GB | — | — |
|
| Qwen3.5 122B-A10B | Won't run | — | Q4_K_M | needs 79.0 GB | — | — |
|
| Solar Open 100B | Won't run | — | Q4_K_M | needs 64.4 GB | — | — |
|
| Solar Open 2 250B | Won't run | — | IQ4_XS | needs 137.2 GB | — | — |
|
Models with a thinking mode spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s). With thinking turned off, a model may do one grade better.
Verdict by memory size
| Model | 16 GB | 24 GB | 32 GB |
|---|---|---|---|
| MiniCPM5 1B | Runs great | Runs great | Runs great |
| Ling-3.0 Tiny | Runs great | Runs great | Runs great |
| Llama 3.2 1B | Runs great | Runs great | Runs great |
| EXAONE 4.0 1.2B | Runs great | Runs great | Runs great |
| Gemma 4 E2B | Runs great | Runs great | Runs great |
| LFM2.5 2.6B | Runs great | Runs great | Runs great |
| HyperCLOVA X SEED 1.5B | Runs great | Runs great | Runs great |
| Mellum2.1 12B-A2.5B Thinking | Runs great | Runs great | Runs great |
| MiniCPM5 2B | Runs great | Runs great | Runs great |
| MiniCPM-V 4.7 35B-A3B | Won't run | Won't run | Runs great |
| gpt-oss-20b | Won't run | Runs great | Runs great |
| Spark-X2.5 4B | Runs great | Runs great | Runs great |
| Llama 3.2 3B | Runs great | Runs great | Runs great |
| Qwen3.5 4B | Runs great | Runs great | Runs great |
| Qwen3 30B-A3B (2507) | Won't run | Runs slowly | Runs great |
| Kanana 1.5 15.7B-A3B | Runs slowly | Runs great | Runs great |
| Gemma 4 E4B | Runs great | Runs great | Runs great |
| Gemma 4 26B-A4B | Won't run | Runs slowly | Runs great |
| Kanana 1.5 8B | Runs well | Runs great | Runs great |
| Llama 3.1 8B | Runs well | Runs great | Runs great |
| DeepSeek R1 Distill Llama 8B | Runs well | Runs well | Runs well |
| Nanbeige4.2 3B | Runs well | Runs well | Runs well |
| Qwen3.5 9B | Runs well | Runs well | Runs well |
| DeepSeek R1 0528 Qwen3 8B | Runs well | Runs well | Runs well |
| Qwen3 8B | Runs well | Runs well | Runs well |
| K2-Horizon 7B | Runs well | Runs well | Runs well |
| Gemma 4 12B | Runs well | Runs well | Runs well |
| Gemma 3 12B | Runs well | Runs well | Runs well |
| HyperCLOVA X SEED Think 14B | Runs slowly | Runs well | Runs well |
| Qwen3 14B | Runs slowly | Runs well | Runs well |
| DeepSeek R1 Distill Qwen 14B | Runs slowly | Runs well | Runs well |
| Phi-4 | Runs slowly | Runs well | Runs well |
| Mistral Small 3.2 24B | Won't run | Runs well | Runs well |
| Qwen3.6 35B-A3B | Won't run | Won't run | Runs slowly |
| Qwen3.5 35B-A3B | Won't run | Won't run | Runs slowly |
| Nemotron 3.5 Lightning 30B-A3B | Won't run | Won't run | Runs slowly |
| K2-Horizon MoVA 36B-A4B | Won't run | Won't run | Runs slowly |
| Gemma 3 27B | Won't run | Runs slowly | Runs slowly |
| Qwen3.6 27B | Won't run | Runs slowly | Runs slowly |
| Muse Glimmer 30B | Won't run | Runs slowly | Runs slowly |
| Qwen3.5 27B | Won't run | Runs slowly | Runs slowly |
| Qwen3.8 27B | Won't run | Runs slowly | Runs slowly |
| Gemma 4 31B | Won't run | Runs slowly | Runs slowly |
| Granite 4.2 30B | Won't run | Runs slowly | Runs slowly |
| EXAONE 4.0 32B | Won't run | Runs slowly | Runs slowly |
| EXAONE 4.5 33B | Won't run | Won't run | Runs slowly |
| Qwen3 32B | Won't run | Won't run | Runs slowly |
| DeepSeek R1 Distill Qwen 32B | Won't run | Won't run | Runs slowly |
| Llama 3.3 70B | Won't run | Won't run | Won't run |
| DeepSeek V4 Flash 0731 | Won't run | Won't run | Won't run |
| GLM-5.3 Flash | Won't run | Won't run | Won't run |
| gpt-oss-120b | Won't run | Won't run | Won't run |
| Kolibri 1 | Won't run | Won't run | Won't run |
| Laguna S 2.1 | Won't run | Won't run | Won't run |
| Ling-3.0 Flash | Won't run | Won't run | Won't run |
| Qwen3.5 122B-A10B | Won't run | Won't run | Won't run |
| Solar Open 100B | Won't run | Won't run | Won't run |
| Solar Open 2 250B | Won't run | Won't run | Won't run |
Measured results on the Apple M6
Public benchmarks we calibrate against. Their conditions (context, backend, flags) can differ from the estimates above.
| Model | Quant | Backend | Context | Prompt (tok/s) | Generation (tok/s) | Flags | Source | Measured |
|---|---|---|---|---|---|---|---|---|
| llama-2-7b | Q4_0 | llama.cpp | 512 | 989.0 | 35.7 | Metal, FA | github.com | 2026-10-01 |
Frequently asked questions
What is the largest model that runs entirely on the Apple M6?
Qwen3.5 35B-A3B at IQ4_XS (a 18.17 GB file) fits entirely in 32 GB of unified memory with 8k context, at est. 42.4 tok/s (33.9–50.9, calibrated estimate ±20%).
How many local LLMs run well on the Apple M6?
At Q4_K_M with 8k context, with 32 GB of memory, out of 58 tracked models: 20 run great, 13 run well, 15 run slowly and 10 won't run.
Can the Apple M6 run a 70B model like Llama 3.3 70B?
Runs slowly — CPU only, IQ2_M (heavy quality loss): est. 3.2 tok/s (2.2–4.1, theoretical estimate ±30%).
Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.