LocalFit
Submit a benchmark

Apple M3 (10-core GPU), 16 GB: which models fit?

RTX 4090RTX 5090RTX 3090RTX 4060 Ti (16 GB)RTX 5070 TiRadeon RX 7900 XTXM4 Max 128 GBM3 Ultra 512 GBDGX Spark (128 GB)All devices

Apple Silicon Mac

Apple M3 (10-core GPU), 16 GB

Memory16 GB unifiedBandwidth100 GB/s vendor specPlanned for models12.0 GiB 75% of unified memoryFilesMLX Apple MLX format

Editorial pick · 16 GB MacsK2-Horizon-7B MLX 8-bit All picks

Context
Files

Budget = file + KV cache + 1 GiB runtime + 0.5 GiB spare. Macs: 75% of memory. Details

19 models fit M3 10c GPU · 16 GB · 8K context

Ranked by Artificial Analysis Intelligence Index, unranked models last. Arena is shown separately. How ranking works

Intelligence

Artificial Analysis Intelligence Index · top 10 models that fit M3 10c GPU · 16 GB · higher is better

  1. K2-Horizon-7B20.6
  2. K2-Horizon-3.7B15.6
  3. Qwen3.5-9B13.3
  4. Qwen3.5-4B13.1
  5. MiniCPM5-2B12.5
  6. Granite 4.2 8B11.1
  7. G9v3-3B10.8
  8. Granite 4.2 3B9.1
  9. MiniCPM5-1B8.8
  10. Nanbeige4.1-3B8.4

AA estimate

Generation speed

Tokens per second on M3 10c GPU · 16 GB, best file for each of the same models · higher is better

  1. MiniCPM5-1B~71.7
  2. G9v3-3B~48.9
  3. MiniCPM5-2B~30.8
  4. Granite 4.2 3B~21.2
  5. Nanbeige4.1-3B~19.7
  6. Qwen3.5-4B~16
  7. K2-Horizon-3.7B~15.3
  8. Granite 4.2 8B~8.8
  9. K2-Horizon-7B~8.6
  10. Qwen3.5-9B~7.9

MeasuredEstimated (~)

WeightsKV cacheRuntimeFreeMeasuredlinks to its sourceEstimatedshown with ~Unknownnever zero
  1. 1

    K2-Horizon-7B

    IFM/K2-Horizon-7B

    2026-09-03size unknown

    AA index20.6rank #33
    Arena—
    Best file & memory
    MLX 8-bit8.9 GiB fileFits
    11.0 of 12.0 GiB8K ctx · max ≈11K
    Prompt—Unknown
    Generate~8.6 t/sEstimated
    First token · 4K prompt—Unknown

    Memory budget

    Weights file8.91 GiBmlx-community/K2-Horizon-7B-8bit · 2 files HF files
    KV cache · 8K1.13 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted11.03 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈11K tokens (native 512K).

    Other MLX files

    • MLX 4-bit 6.7 GiB · fits · files

    Where the speeds come from

    • Prompt unknown: no measured report, and no comparable vendor FP16 figure or device calibration.
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 8.9 GiB read per token.
    • First token unknown: needs a prefill value.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 20.6 K2 Horizon 7B · Arena unknown · Model page

  2. 2

    K2-Horizon-3.7B

    IFM/K2-Horizon-3.7B

    2026-09-03size unknown

    AA index15.6rank #41
    Arena—
    Best file & memory
    MLX 8-bit5.0 GiB fileFits
    7.1 of 12.0 GiB8K ctx · max ≈39K
    Prompt—Unknown
    Generate~15.3 t/sEstimated
    First token · 4K prompt—Unknown

    Memory budget

    Weights file5.01 GiBmlx-community/K2-Horizon-3.7B-8bit · 2 files HF files
    KV cache · 8K1.13 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted7.13 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈39K tokens (native 512K).

    Other MLX files

    • MLX 4-bit 3.9 GiB · fits · files

    Where the speeds come from

    • Prompt unknown: no measured report, and no comparable vendor FP16 figure or device calibration.
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 5.0 GiB read per token.
    • First token unknown: needs a prefill value.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 15.6 K2 Horizon 3.7B · Arena unknown · Model page

  3. 3

    Qwen3.5-9B

    Qwen/Qwen3.5-9B

    2026-03-029B dense

    AA index13.3 AA estimaterank #45
    Arena—
    Best file & memory
    MLX 8-bit9.7 GiB fileFits · conditional
    11.0 of 12.0 GiB+8K ctx · max ≈25K
    Prompt~165 t/sEstimated
    Generate~7.9 t/sEstimated
    First token · 4K prompt~25 sEstimated

    Memory budget

    Weights file9.71 GiBmlx-community/Qwen3.5-9B-8bit · 2 files HF files
    KV cache · 8K0.25 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted10.96 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈25K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 5.5 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 9.7 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 13.3 AA estimate Qwen3.5 9B (Non-reasoning) · Arena unknown · Model page

  4. 4

    Qwen3.5-4B

    Qwen/Qwen3.5-4B

    2026-03-024B dense

    AA index13.1 AA estimaterank #48
    Arena—
    Best file & memory
    MLX 8-bit4.8 GiB fileFits · conditional
    6.0 of 12.0 GiB+8K ctx · max ≈183K
    Prompt~370 t/sEstimated
    Generate~16 t/sEstimated
    First token · 4K prompt~11 sEstimated

    Memory budget

    Weights file4.78 GiBmlx-community/Qwen3.5-4B-8bit HF files
    KV cache · 8K0.25 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted6.03 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈183K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 2.8 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 4.8 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 13.1 AA estimate Qwen3.5 4B (Reasoning) · Arena unknown · Model page

  5. 5

    MiniCPM5-2B

    openbmb/MiniCPM5-2B

    2026-09-072.5B dense

    AA index12.5rank #51
    Arena—
    Best file & memory
    MLX 8-bit2.5 GiB fileFits
    3.8 of 12.0 GiB8K ctx · max ≈128K
    Prompt~589 t/sEstimated
    Generate~30.8 t/sEstimated
    First token · 4K prompt~7.0 sEstimated

    Memory budget

    Weights file2.49 GiBmlx-community/MiniCPM5-2B-8bit HF files
    KV cache · 8K0.33 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted3.82 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈128K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 1.3 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 2.5 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 12.5 MiniCPM5-2B · Arena unknown · Model page

  6. 6

    Granite 4.2 8B

    ibm-granite/granite-4.2-8b

    2026-08-258B dense

    AA index11.1rank #56
    Arena1,320
    Best file & memory
    MLX 8-bit8.7 GiB fileFits
    10.9 of 12.0 GiB8K ctx · max ≈12K
    Prompt~185 t/sEstimated
    Generate~8.8 t/sEstimated
    First token · 4K prompt~22 sEstimated

    Memory budget

    Weights file8.70 GiBlmstudio-community/granite-4.2-8b-MLX-8bit · 2 files HF files
    KV cache · 8K1.25 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted10.95 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈12K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 4.6 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 8.7 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 11.1 Granite 4.2 8B · Arena 1,320 · Model page

  7. 7

    G9v3-3B

    AI9Stars/G9v3-3B

    2026-07 (day unknown)3B dense

    AA index10.8 AA estimaterank #58
    Arena—
    Best file & memory
    MLX 4-bit1.6 GiB fileFits
    3.0 of 12.0 GiB8K ctx · max ≈128K
    Prompt~494 t/sEstimated
    Generate~48.9 t/sEstimated
    First token · 4K prompt~8.3 sEstimated

    Memory budget

    Weights file1.57 GiBcof139/G9v3-3B-mlx-4Bit HF files
    KV cache · 8K0.41 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted2.97 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈128K tokens (native 128K).

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 1.6 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 10.8 AA estimate G9v3-3B · Arena unknown · Model page

  8. 8

    Granite 4.2 3B

    ibm-granite/granite-4.2-3b

    2026-08-253B dense

    AA index9.1rank #64
    Arena1,293
    Best file & memory
    MLX 8-bit3.6 GiB fileFits
    5.2 of 12.0 GiB8K ctx · max ≈88K
    Prompt~494 t/sEstimated
    Generate~21.2 t/sEstimated
    First token · 4K prompt~8.3 sEstimated

    Memory budget

    Weights file3.62 GiBlmstudio-community/granite-4.2-3b-MLX-8bit HF files
    KV cache · 8K0.63 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted5.25 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈88K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 1.9 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 3.6 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 9.1 Granite 4.2 3B · Arena 1,293 · Model page

  9. 9

    MiniCPM5-1B

    openbmb/MiniCPM5-1B

    2026-05-191.1B dense

    AA index8.8 AA estimaterank #67
    Arena—
    Best file & memory
    MLX 8-bit1.1 GiB fileFits
    2.3 of 12.0 GiB8K ctx · max ≈128K
    Prompt~1,371 t/sEstimated
    Generate~71.7 t/sEstimated
    First token · 4K prompt~3.0 sEstimated

    Memory budget

    Weights file1.07 GiBmlx-community/MiniCPM5-1B-8bit HF files
    KV cache · 8K0.19 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted2.26 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈128K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 0.6 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 1.1 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 8.8 AA estimate MiniCPM5-1B (Reasoning) · Arena unknown · Model page

  10. 10

    Nanbeige4.1-3B

    Nanbeige/Nanbeige4.1-3B

    2026-02 (day unknown)3B dense

    AA index8.4 AA estimaterank #69
    Arena—
    Best file & memory
    MLX 8-bit3.9 GiB fileFits
    5.4 of 12.0 GiB8K ctx · max ≈106K
    Prompt~494 t/sEstimated
    Generate~19.7 t/sEstimated
    First token · 4K prompt~8.3 sEstimated

    Memory budget

    Weights file3.89 GiBmlx-community/Nanbeige4.1-3B-8bit HF files
    KV cache · 8K0.50 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted5.39 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈106K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 2.1 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 3.9 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 8.4 AA estimate Nanbeige4.1-3B · Arena unknown · Model page

  11. 11

    LFM2.5-2.6B

    LiquidAI/LFM2.5-2.6B

    2026-08-042.7B dense

    AA index8.4 AA estimaterank #70
    Arena—
    Best file & memory
    MLX 8-bit2.7 GiB fileFits · conditional
    3.8 of 12.0 GiB+8K ctx · max ≈128K
    Prompt~551 t/sEstimated
    Generate~28.7 t/sEstimated
    First token · 4K prompt~7.4 sEstimated

    Memory budget

    Weights file2.67 GiBLiquidAI/LFM2.5-2.6B-MLX-8bit HF files
    KV cache · 8K0.13 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted3.79 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈128K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 1.5 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 2.7 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 8.4 AA estimate LFM2.5-2.6B · Arena unknown · Model page

  12. 12

    Falcon-H1R-7B

    tiiuae/Falcon-H1R-7B

    2026-01-057B dense

    AA index7.8 AA estimaterank #71
    Arena—
    Best file & memory
    MLX 8-bit7.5 GiB fileFits · conditional
    8.9 of 12.0 GiB+8K ctx · max ≈70K
    Prompt~212 t/sEstimated
    Generate~10.2 t/sEstimated
    First token · 4K prompt~19 sEstimated

    Memory budget

    Weights file7.51 GiBmlx-community/Falcon-H1R-7B-8bit · 2 files HF files
    KV cache · 8K0.34 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted8.85 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈70K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 4.0 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 7.5 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 7.8 AA estimate Falcon-H1R-7B · Arena unknown · Model page

  13. 13

    LFM2.5-8B-A1B

    LiquidAI/LFM2.5-8B-A1B

    2026-05-288.3B · 1.5B active

    AA index7.2 AA estimaterank #74
    Arena—
    Best file & memory
    MLX 8-bit8.4 GiB fileFits · conditional
    9.5 of 12.0 GiB+8K ctx · max ≈125K
    Prompt~670 t/sEstimated
    Generate~21 t/sEstimated
    First token · 4K prompt~6.1 sEstimated

    Memory budget

    Weights file8.38 GiBLiquidAI/LFM2.5-8B-A1B-MLX-8bit · 2 files HF files
    KV cache · 8K0.09 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted9.48 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈125K tokens (native 125K).

    Other MLX files

    • MLX 4-bit 4.5 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 52 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.3416 (n=94) × memory bandwidth ÷ 1.5 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 7.2 AA estimate LFM2.5-8B-A1B · Arena unknown · Model page

  14. 14

    NVIDIA Nemotron 3 Nano 4B

    nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

    2026-03-164B dense

    AA index4.8rank #75
    Arena—
    Best file & memory
    MLX 8-bit3.9 GiB fileFits · conditional
    5.1 of 12.0 GiB+8K ctx · max ≈256K
    Prompt~373 t/sEstimated
    Generate~19.5 t/sEstimated
    First token · 4K prompt~11 sEstimated

    Memory budget

    Weights file3.93 GiBmlx-community/NVIDIA-Nemotron-3-Nano-4B-8bit HF files
    KV cache · 8K0.13 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted5.06 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈256K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 2.1 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 3.9 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 4.8 NVIDIA Nemotron 3 Nano 4B · Arena unknown · Model page

  15. 15

    MiMo-V2.6-Distill-Qwen-9B

    XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B

    2026-09-229B dense

    AA index—unranked
    Arena—
    Best file & memory
    MLX 8-bit9.7 GiB fileFits · conditional
    11.0 of 12.0 GiB+8K ctx · max ≈25K
    Prompt~165 t/sEstimated
    Generate~7.9 t/sEstimated
    First token · 4K prompt~25 sEstimated

    Memory budget

    Weights file9.71 GiBnativ-community/MiMo-V2.6-Distill-Qwen-9B-MLX-8bit · 2 files HF files
    KV cache · 8K0.25 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted10.96 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈25K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 5.5 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 9.7 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA index unknown · Arena unknown · Model page

  16. 16

    K2-Horizon-0.9B

    IFM/K2-Horizon-0.9B

    2026-09-03size unknown

    AA index—unranked
    Arena—
    Best file & memory
    MLX 8-bit1.1 GiB fileFits
    2.5 of 12.0 GiB8K ctx · max ≈128K
    Prompt—Unknown
    Generate~71.8 t/sEstimated
    First token · 4K prompt—Unknown

    Memory budget

    Weights file1.07 GiBmlx-community/K2-Horizon-0.9B-8bit HF files
    KV cache · 8K0.44 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Accounted2.50 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈128K tokens (native 128K).

    Other MLX files

    • MLX 4-bit 0.8 GiB · fits · files

    Where the speeds come from

    • Prompt unknown: no measured report, and no comparable vendor FP16 figure or device calibration.
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 1.1 GiB read per token.
    • First token unknown: needs a prefill value.

    Caveats

    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA index unknown · Arena unknown · Model page

  17. 17

    Ornith-1.5-9B

    ornith-ai/Ornith-1.5-9B

    2026-08 (day unknown)9B dense

    AA index—unranked
    Arena—
    Best file & memory
    MLX 8-bit8.9 GiB fileFits · conditional
    10.1 of 12.0 GiB+8K ctx · max ≈52K
    Prompt~165 t/sEstimated
    Generate~8.7 t/sEstimated
    First token · 4K prompt~25 sEstimated

    Memory budget

    Weights file8.86 GiBornith-ai/Ornith-1.5-9B-MLX-8bit · 2 files HF files
    KV cache · 8K0.25 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted10.11 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈52K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 4.7 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 8.9 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA index unknown · Arena unknown · Model page

  18. 18

    Qwen3.5-0.8B

    Qwen/Qwen3.5-0.8B

    2026-03-020.8B dense

    AA index—unranked
    Arena—
    Best file & memory
    MLX 8-bit0.9 GiB fileFits · conditional
    2.0 of 12.0 GiB+8K ctx · max ≈256K
    Prompt~1,852 t/sEstimated
    Generate~82.2 t/sEstimated
    First token · 4K prompt~2.2 sEstimated

    Memory budget

    Weights file0.93 GiBmlx-community/Qwen3.5-0.8B-8bit HF files
    KV cache · 8K0.09 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted2.03 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈256K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 0.6 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 0.9 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA index unknown · Arena unknown · Model page

  19. 19

    Qwen3.5-2B

    Qwen/Qwen3.5-2B

    2026-03-022B dense

    AA index—unranked
    Arena—
    Best file & memory
    MLX 8-bit2.5 GiB fileFits · conditional
    3.6 of 12.0 GiB+8K ctx · max ≈256K
    Prompt~741 t/sEstimated
    Generate~30.9 t/sEstimated
    First token · 4K prompt~5.5 sEstimated

    Memory budget

    Weights file2.48 GiBmlx-community/Qwen3.5-2B-8bit HF files
    KV cache · 8K0.09 GiBFP16 cache from config.json Estimated
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted3.57 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: ≈256K tokens (native 256K).

    Other MLX files

    • MLX 4-bit 1.6 GiB · fits · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 2.5 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA index unknown · Arena unknown · Model page

May fit: KV-cache size unknown 4

Weights and runtime fit, but the cache for this attention type is not established. Ranked by AA index.

  1. Gemma 4 12B (it)

    google/gemma-4-12B-it

    2026-06-0312B dense

    AA index14.2 AA estimaterank #44
    Arena—
    Best file & memory
    MLX 4-bit6.3 GiB fileCache unknown
    7.3 of 12.0 GiB+8K ctx
    Prompt~124 t/sEstimated
    Generate~12.2 t/sEstimated
    First token · 4K prompt~33 sEstimated

    Memory budget

    Weights file6.28 GiBmlx-community/gemma-4-12B-it-4bit · 2 files HF files
    KV cache · 8K—Unknown
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted7.28 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: unknown (native 256K).

    Other MLX files

    • MLX 8-bit 11.8 GiB · too big · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 6.3 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • KV-cache size could not be established from the config (compressed or backend-specific attention).
    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 14.2 AA estimate Gemma 4 12B (Reasoning) · Arena unknown · Model page

  2. Ling-3.0-tiny

    inclusionAI/Ling-3.0-tiny

    2026-08 (day unknown)7.9B · 1.3B active

    AA index11.1rank #57
    Arena—
    Best file & memory
    MLX 4-bit4.1 GiB fileCache unknown
    5.1 of 12.0 GiB+8K ctx
    Prompt~773 t/sEstimated
    Generate~46.7 t/sEstimated
    First token · 4K prompt~5.3 sEstimated

    Memory budget

    Weights file4.14 GiBrapid-mlx/Ling-3.0-tiny-MLX-4bit HF files
    KV cache · 8K—Unknown
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted5.14 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: unknown (native 128K).

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 52 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.3416 (n=94) × memory bandwidth ÷ 0.7 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • KV-cache size could not be established from the config (compressed or backend-specific attention).
    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 11.1 Ling 3.0 Tiny · Arena unknown · Model page

  3. Gemma 4 E4B (it)

    google/gemma-4-E4B-it

    2026-04-028B dense

    AA index8.9rank #66
    Arena—
    Best file & memory
    MLX 8-bit8.3 GiB fileCache unknown
    9.3 of 12.0 GiB+8K ctx
    Prompt~185 t/sEstimated
    Generate~9.3 t/sEstimated
    First token · 4K prompt~22 sEstimated

    Memory budget

    Weights file8.27 GiBmlx-community/gemma-4-E4B-it-8bit · 2 files HF files
    KV cache · 8K—Unknown
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted9.27 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: unknown (native 128K).

    Other MLX files

    • MLX 4-bit 4.8 GiB · cache unknown · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 8.3 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • KV-cache size could not be established from the config (compressed or backend-specific attention).
    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 8.9 Gemma 4 E4B (Reasoning) · Arena unknown · Model page

  4. Gemma 4 E2B

    google/gemma-4-E2B-it

    2026-04-025.1B dense

    AA index7.8 AA estimaterank #72
    Arena—
    Best file & memory
    MLX 8-bit5.5 GiB fileCache unknown
    6.5 of 12.0 GiB+8K ctx
    Prompt~290 t/sEstimated
    Generate~14 t/sEstimated
    First token · 4K prompt~14 sEstimated

    Memory budget

    Weights file5.46 GiBmlx-community/gemma-4-E2B-it-8bit · 2 files HF files
    KV cache · 8K—Unknown
    Runtime allowance1.00 GiBplanning assumption Estimated
    Recurrent state—Unknown
    Accounted6.46 GiBof 12.0 GiB usable (75% of 16 GB) · 0.5 GiB spare margin required

    Max context in free memory: unknown (native 128K).

    Other MLX files

    • MLX 4-bit 3.3 GiB · cache unknown · files

    Where the speeds come from

    • Prompt estimated from median prefill rate per GPU core of 8 measured M3 points (MLX) × this chip's GPU cores ÷ (2 × active params).
    • Generate estimated from MLX median efficiency 0.8231 (n=60) × memory bandwidth ÷ 5.5 GiB read per token.
    • First token 4K-token prompt ÷ estimated prefill speed. Excludes model load and queueing.

    Caveats

    • KV-cache size could not be established from the config (compressed or backend-specific attention).
    • Fixed recurrent/convolution state is unknown and not included.
    • Text-only budget. A downloadable file does not prove your runtime supports this architecture.

    AA 7.8 AA estimate Gemma 4 E2B (Reasoning) · Arena unknown · Model page

48 models too big for M3 10c GPU · 16 GB

Specs & sources

Memory16 GB unifiedsource
Bandwidth100 GB/ssource
FP16unknownApple publishes no comparable FP16 throughput or GPU-only TDP in reviewed chip/device technical specifications. Power adapter rating and whole-system maximum continuous power are not GPU TDP.
PowerunknownApple publishes no comparable FP16 throughput or GPU-only TDP in reviewed chip/device technical specifications. Power adapter rating and whole-system maximum continuous power are not GPU TDP.

Found in: MacBook Air (13-inch, M3, 2024)

Measured speeds

2 reported observations

On this device (same chip and GPU-core count). Reported by the linked sources; not reproduced by LocalFit.

Open in speed databaseSubmit yours

Doesn't fit? Rent a GPU

  • RunpodPlaceholderCloud GPU pods you can rent for running larger models.
  • Vast.aiPlaceholderGPU rental marketplace with many hosts and GPU types.
  • LambdaPlaceholderCloud GPU instances for AI training and inference.
  • TensorDockPlaceholderGPU cloud marketplace with consumer and datacenter GPUs.

Sponsored placeholder links: LocalFit may earn a commission if these become affiliate links. They never influence fit results, estimates or picks.