Apple Silicon (Mac)
Apple M5
Data as of
- Memory bandwidth
- 153 GB/s
- Source
- Checked:
With 16 GB
One entry per model, up to 30, sorted by popularity. For each model we show the largest (most precise) quantization that fits in this memory with a context of at least 8K tokens. The context length is the longest one that quantization fits (for models whose maximum is under 8K, such as embedding models, that maximum). Models that are not for chat or text generation are labelled with their use. "Left" is the memory minus the estimate; under 5% of the memory is marked "tight" (not a guarantee that it fits).
-
Qwen3.8-27B UD-Q4_K_SContext length 8,192 Estimated 17.1 GB 0.08 GB left (tight) Speed 6.0–8.0
-
Ornith-1.5-9B Q8_0Context length 131,072 Estimated 15.2 GB 1.97 GB left Speed 9.4–13
-
Context length 131,072 Estimated 17.0 GB 0.14 GB left (tight) Speed not estimated (MoE)
-
Context length 8,192 Estimated 17.2 GB 0.02 GB left (tight) Speed 6.0–7.9
-
Context length 65,536 Estimated 16.5 GB 0.66 GB left (tight) Speed not estimated (MoE)
-
Context length 65,536 Estimated 16.3 GB 0.89 GB left Speed 6.7–9.0
-
Qwen3.8-27B-Uncensored IQ4_XSContext length 8,192 Estimated 17.0 GB 0.13 GB left (tight) Speed 6.0–8.0
-
Ornith-1.0-9B Q8_0Context length 131,072 Estimated 14.9 GB 2.23 GB left Speed 9.6–13
-
Context length 512 Estimated 2.41 GB 14.8 GB left Speed 69–92
-
Context length 32,768 Estimated 17.1 GB 0.05 GB left (tight) Speed not estimated (MoE)
-
Context length 131,072 Estimated 16.1 GB 1.05 GB left Speed 8.5–11
-
Context length 40,960 Estimated 15.6 GB 1.58 GB left Speed 11–14
-
Context length 32,768 Estimated 16.8 GB 0.39 GB left (tight) Speed 6.8–9.1
-
Context length 32,768 Estimated 16.8 GB 0.36 GB left (tight) Speed not estimated (MoE)
-
Context length 32,768 Estimated 16.5 GB 0.68 GB left (tight) Speed 8.5–11
-
Context length 65,536 Estimated 16.2 GB 0.96 GB left Speed 7.1–9.4
-
Context length 131,072 Estimated 14.9 GB 2.26 GB left Speed 9.6–13
-
Context length 8,192 Estimated 17.0 GB 0.14 GB left (tight) Speed not estimated (MoE)
-
Ternary-Bonsai-2-27B Q2_0Context length 65,536 Estimated 13.1 GB 4.06 GB left Speed 12–16
-
Context length 32,768 Estimated 15.5 GB 1.71 GB left Speed 7.6–10
-
Context length 131,072 Estimated 14.1 GB 3.12 GB left Speed 11–14
-
Context length 8,192 Estimated 17.0 GB 0.13 GB left (tight) Speed 6.0–8.0
-
Context length 8,192 Estimated 17.1 GB 0.08 GB left (tight) Speed 6.0–8.0
-
Wan2.1 IQ3_SContext length 32,768 Estimated 16.0 GB 1.21 GB left Speed 7.3–9.7
-
Context length 32,768 Estimated 15.5 GB 1.71 GB left Speed 7.6–10
-
Context length 65,536 Estimated 12.5 GB 4.72 GB left Speed 22–30
-
Context length 65,536 Estimated 16.3 GB 0.89 GB left Speed not estimated (MoE)
-
Context length 16,384 Estimated 16.9 GB 0.29 GB left (tight) Speed not estimated (MoE)
-
Context length 131,072 Estimated 8.42 GB 8.76 GB left Speed 18–24
-
Context length 262,144 Estimated 12.4 GB 4.82 GB left Speed 34–46
With 24 GB
One entry per model, up to 30, sorted by popularity. For each model we show the largest (most precise) quantization that fits in this memory with a context of at least 8K tokens. The context length is the longest one that quantization fits (for models whose maximum is under 8K, such as embedding models, that maximum). Models that are not for chat or text generation are labelled with their use. "Left" is the memory minus the estimate; under 5% of the memory is marked "tight" (not a guarantee that it fits).
-
Context length 8,192 Estimated 25.6 GB 0.17 GB left (tight) Speed 3.8–5.1
-
Ornith-1.5-9B BF16Context length 131,072 Estimated 23.8 GB 1.94 GB left Speed 5.0–6.6
-
Context length 131,072 Estimated 25.5 GB 0.26 GB left (tight) Speed not estimated (MoE)
-
Context length 65,536 Estimated 24.9 GB 0.87 GB left (tight) Speed not estimated (MoE)
-
Context length 8,192 Estimated 25.7 GB 0.09 GB left (tight) Speed 3.8–5.1
-
Context length 262,144 Estimated 23.6 GB 2.22 GB left Speed not estimated (MoE)
-
Context length 65,536 Estimated 22.6 GB 3.15 GB left Speed 5.4–7.1
-
Context length 65,536 Estimated 25.2 GB 0.57 GB left (tight) Speed not estimated (MoE)
-
Context length 16,384 Estimated 25.7 GB 0.09 GB left (tight) Speed 3.9–5.1
-
Context length 16,384 Estimated 24.7 GB 1.06 GB left (tight) Speed 4.1–5.5
-
Ornith-1.0-9B BF16Context length 131,072 Estimated 23.3 GB 2.43 GB left Speed 5.1–6.8
-
Context length 512 Estimated 2.41 GB 23.4 GB left Speed 69–92
-
Context length 65,536 Estimated 25.3 GB 0.46 GB left (tight) Speed not estimated (MoE)
-
Context length 131,072 Estimated 23.8 GB 1.97 GB left Speed 5.0–6.6
-
Context length 40,960 Estimated 23.3 GB 2.49 GB left Speed 5.6–7.5
-
Context length 65,536 Estimated 25.7 GB 0.05 GB left (tight) Speed 4.5–6.1
-
Context length 32,768 Estimated 25.4 GB 0.42 GB left (tight) Speed not estimated (MoE)
-
Context length 32,768 Estimated 22.1 GB 3.70 GB left Speed 5.6–7.5
-
Context length 131,072 Estimated 23.8 GB 1.97 GB left Speed 5.0–6.6
-
Context length 131,072 Estimated 23.3 GB 2.46 GB left Speed 5.1–6.8
-
Context length 65,536 Estimated 25.3 GB 0.45 GB left (tight) Speed not estimated (MoE)
-
Ternary-Bonsai-2-27B Q2_0Context length 131,072 Estimated 17.4 GB 8.35 GB left Speed 12–16
-
Context length 131,072 Estimated 21.9 GB 3.86 GB left Speed 7.6–10
-
Context length 262,144 Estimated 18.4 GB 7.42 GB left Speed 11–14
-
Ornith-1.0-35B Q4_K_MContext length 131,072 Estimated 25.0 GB 0.78 GB left (tight) Speed not estimated (MoE)
-
Context length 8,192 Estimated 25.3 GB 0.45 GB left (tight) Speed 3.9–5.2
-
Context length 32,768 Estimated 25.3 GB 0.44 GB left (tight) Speed 4.2–5.6
-
Wan2.1 Q4_K_MContext length 65,536 Estimated 22.3 GB 3.43 GB left Speed 5.5–7.3
-
Context length 131,072 Estimated 21.9 GB 3.86 GB left Speed 7.6–10
-
Context length 65,536 Estimated 12.5 GB 13.3 GB left Speed 22–30
With 32 GB
One entry per model, up to 30, sorted by popularity. For each model we show the largest (most precise) quantization that fits in this memory with a context of at least 8K tokens. The context length is the longest one that quantization fits (for models whose maximum is under 8K, such as embedding models, that maximum). Models that are not for chat or text generation are labelled with their use. "Left" is the memory minus the estimate; under 5% of the memory is marked "tight" (not a guarantee that it fits).
-
Qwen3.8-27B UD-Q8_K_XLContext length 16,384 Estimated 33.7 GB 0.63 GB left (tight) Speed 2.9–3.9
-
Ornith-1.5-9B BF16Context length 262,144 Estimated 28.1 GB 6.24 GB left Speed 5.0–6.6
-
Context length 131,072 Estimated 33.0 GB 1.36 GB left (tight) Speed not estimated (MoE)
-
Context length 65,536 Estimated 33.2 GB 1.13 GB left (tight) Speed not estimated (MoE)
-
Context length 16,384 Estimated 33.7 GB 0.63 GB left (tight) Speed 2.9–3.9
-
Context length 262,144 Estimated 33.6 GB 0.74 GB left (tight) Speed not estimated (MoE)
-
Context length 131,072 Estimated 26.9 GB 7.44 GB left Speed 5.4–7.1
-
Context length 32,768 Estimated 34.0 GB 0.34 GB left (tight) Speed not estimated (MoE)
-
Context length 262,144 Estimated 29.7 GB 4.66 GB left Speed 3.9–5.1
-
Context length 32,768 Estimated 32.4 GB 1.96 GB left Speed 3.2–4.2
-
Ornith-1.0-9B BF16Context length 262,144 Estimated 27.6 GB 6.72 GB left Speed 5.1–6.8
-
Context length 512 Estimated 2.41 GB 31.9 GB left Speed 69–92
-
Context length 16,384 Estimated 34.1 GB 0.30 GB left (tight) Speed not estimated (MoE)
-
Context length 262,144 Estimated 28.1 GB 6.26 GB left Speed 5.0–6.6
-
Context length 40,960 Estimated 23.3 GB 11.1 GB left Speed 5.6–7.5
-
Context length 16,384 Estimated 33.7 GB 0.63 GB left (tight) Speed 2.9–3.9
-
Context length 131,072 Estimated 31.7 GB 2.65 GB left Speed not estimated (MoE)
-
Context length 65,536 Estimated 26.9 GB 7.46 GB left Speed 5.6–7.5
-
Context length 262,144 Estimated 28.1 GB 6.26 GB left Speed 5.0–6.6
-
Context length 262,144 Estimated 27.6 GB 6.75 GB left Speed 5.1–6.8
-
Context length 65,536 Estimated 34.3 GB 0.06 GB left (tight) Speed not estimated (MoE)
-
Ternary-Bonsai-2-27B Q2_0Context length 262,144 Estimated 26.0 GB 8.35 GB left Speed 12–16
-
Context length 262,144 Estimated 30.5 GB 3.86 GB left Speed 7.6–10
-
Context length 262,144 Estimated 18.4 GB 16.0 GB left Speed 11–14
-
Ornith-1.0-35B Q6_KContext length 131,072 Estimated 32.3 GB 2.02 GB left Speed not estimated (MoE)
-
Context length 32,768 Estimated 33.6 GB 0.77 GB left (tight) Speed 3.0–4.0
-
Context length 16,384 Estimated 33.7 GB 0.63 GB left (tight) Speed 2.9–3.9
-
Wan2.1 IQ4_NLContext length 65,536 Estimated 34.3 GB 0.03 GB left (tight) Speed 3.2–4.2
-
Context length 262,144 Estimated 30.5 GB 3.86 GB left Speed 7.6–10
-
Context length 65,536 Estimated 12.5 GB 21.9 GB left Speed 22–30
Frequently asked questions
- Which local LLMs run on Apple M5?
- On this site, 965 models have a quantization that fits in 32 GB at 8,192 tokens (estimated). Popular ones include Qwen3.8-27B, Ornith-1.5-9B, Huihui-Qwen3.6-35B-A3B-abliterated-MTP, Ornith-1.5-35B-A3B, Huihui-Qwen3.8-27B-abliterated.
- What is the largest model that runs on Apple M5?
- Among the models listed here, Mixtral-8x22B-Instruct-v0.1 (about 140.6B); IQ1_S fits in 32 GB (estimated at 8,192 tokens).
- How fast does Apple M5 generate text?
- With 153 GB/s of memory bandwidth, Qwen3.8-27B at UD-Q8_K_XL (31.5 GB) runs at an estimated 2.9–3.9 tokens per second (60–80% of bandwidth ÷ file size; prompt processing not included).