Skip to content
AI Atlas
Hardware Fit tool

What fits my machine?

Pick a device memory size, a quantization and a context window: the atlas estimates each model's memory need from its parameter count. Every figure here is an estimate, never a measurement.

Estimated

The custom field wins over the preset when both are filled. Estimates reserve 2 GB for the OS and framework.

EstimatedAll figures are estimates, not measurements.192 GB · fp16 / bf16 · 128k tokens552 of 658 models with a known parameter count fit

  • Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead.
  • bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
  • KV cache approximated at 0.5 GB per 8 192 tokens of context, independent of architecture (GQA/MLA models need less).
  • A model 'fits' when the estimate is at most the device memory minus 2 GB reserved for the OS and framework.
  • Mixture-of-experts models are estimated on total parameters (all experts must be resident); active parameters are ignored.
  • Device memory uses the largest configuration when several are listed (e.g. Apple silicon tiers).

Show only fitting modelsMethod →

Estimated hardware fit
ModelParamsQuantEst. memoryHeadroomFitsCompare
Intel/Qwen3.8-Flash-Next-W4A16-AutoRoundOpen weightsIntelestimated: 75.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens75.4Bfp16181.3 GB+8.7 GB✓ fits
UI-TARS-72B-DPOOpen weightsByteDanceestimated: 73.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens73.4Bfp16176.8 GB+13.2 GB✓ fits
Kimi-Dev-72BOpen weightsMoonshot AIestimated: 72.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens72.7Bfp16175.2 GB+14.8 GB✓ fits
amd/Llama-3.3-70B-Instruct-FP8-KVOpen weightsAMDestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
fireworks-ai/llama-3-firefunction-v2Open weightsFireworks AIestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
NousResearch/Meta-Llama-3-70B-InstructOpen weightsNous Researchestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
NousResearch/Meta-Llama-3.1-70B-InstructOpen weightsNous Researchestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
Hermes-4-70BOpen weightsNous Researchestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
Meta-Llama-3-70BRestrictedMeta AIestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
Llama 3.3 70B InstructRestrictedMeta AIestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
Llama-3.1-70B-InstructRestrictedMeta AIestimated: 70.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens70.5Bfp16170.3 GB+19.7 GB✓ fits
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4Open weightsNVIDIAestimated: 67.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens67.2Bfp16162.6 GB+27.4 GB✓ fits
nvidia/Qwen3.5-122B-A10B-NVFP4Open weightsNVIDIAestimated: 64.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens64.6Bfp16156.5 GB+33.5 GB✓ fits
amd/gpt-oss-120b-w-mxfp4-a-fp8Open weightsAMDestimated: 59.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens59.5Bfp16144.8 GB+45.1 GB✓ fits
ai21labs/AI21-Jamba-Mini-1.7-FP8RestrictedAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
ai21labs/AI21-Jamba2-Mini-FP8Open weightsAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
AI21-Jamba-Mini-1.6RestrictedAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
Jamba-v0.1Open weightsAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
AI21-Jamba2-MiniOpen weightsAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
AI21-Jamba-Mini-1.7RestrictedAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
AI21-Jamba-Mini-1.5RestrictedAI21 Labsestimated: 51.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens51.6Bfp16126.6 GB+63.4 GB✓ fits
Kimi-Linear-48B-A3B-BaseOpen weightsMoonshot AIestimated: 49.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens49.1Bfp16121.0 GB+69.0 GB✓ fits
Kimi-Linear-48B-A3B-InstructOpen weightsMoonshot AIestimated: 49.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens49.1Bfp16121.0 GB+69.0 GB✓ fits
function-calling-v1Open weightsFireworks AIestimated: 46.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens46.7Bfp16115.4 GB+74.6 GB✓ fits
Nous-Hermes-2-Mixtral-8x7B-DPOOpen weightsNous Researchestimated: 46.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens46.7Bfp16115.4 GB+74.6 GB✓ fits
firefunction-v1Open weightsFireworks AIestimated: 46.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens46.7Bfp16115.4 GB+74.6 GB✓ fits
Mixtral-8x7B-Instruct-v0.1Open weightsMistral AIestimated: 46.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens46.7Bfp16115.4 GB+74.6 GB✓ fits
DFN2B-CLIP-ViT-L-14-39BOpen weightsAppleestimated: 39.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens39Bfp1697.7 GB+92.3 GB✓ fits
Seed-OSS-36B-BaseOpen weightsByteDanceestimated: 36.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens36.1Bfp1691.2 GB+98.8 GB✓ fits
NousResearch/Hermes-4.3-36B-GGUFOpen weightsNous Researchestimated: 36.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens36.1Bfp1691.2 GB+98.8 GB✓ fits
Seed-OSS-36B-InstructOpen weightsByteDanceestimated: 36.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens36.1Bfp1691.2 GB+98.8 GB✓ fits
Qwen/Qwen3.6-35B-A3B-FP8Open weightsQwenestimated: 36.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens36Bfp1690.7 GB+99.3 GB✓ fits
Qwen3.6 35B A3BOpen weightsQwenestimated: 36.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens36Bfp1690.7 GB+99.3 GB✓ fits
bartowski/endless-frontier_BigBang-v1-GGUFOpen weightsbartowskiestimated: 35.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens35.5Bfp1689.7 GB+100.3 GB✓ fits
unsloth/Qwen3.6-35B-A3B-MTP-GGUFOpen weightsUnslothestimated: 35.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens35.5Bfp1689.7 GB+100.3 GB✓ fits
cerebras/Kimi-Linear-REAP-35B-A3B-InstructOpen weightsCerebras Systemsestimated: 35.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens35.1Bfp1688.8 GB+101.2 GB✓ fits
perplexity-ai/pplx-computer-qwen-3-6-35b-a3b-nvfp4-20260709Open weightsPerplexity AIestimated: 35.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens35.1Bfp1688.8 GB+101.3 GB✓ fits
bartowski/Qwen_Qwen3.6-35B-A3B-GGUFOpen weightsbartowskiestimated: 35.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens35Bfp1688.5 GB+101.5 GB✓ fits
bartowski/Qwen_Qwen3.5-35B-A3B-GGUFOpen weightsbartowskiestimated: 35.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens35Bfp1688.5 GB+101.5 GB✓ fits
c4ai-command-r-v01RestrictedCohereestimated: 35.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens35Bfp1688.5 GB+101.5 GB✓ fits
unsloth/Qwen3.6-35B-A3B-GGUFOpen weightsUnslothestimated: 34.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.7Bfp1687.7 GB+102.3 GB✓ fits
bartowski/XYZAILab_XYZ-Aquila-mini-GGUFOpen weightsbartowskiestimated: 34.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.7Bfp1687.7 GB+102.3 GB✓ fits
bartowski/thomsonreuters_Thomson-1.0-Small-GGUFOpen weightsbartowskiestimated: 34.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.7Bfp1687.7 GB+102.3 GB✓ fits
bartowski/Kwaipilot_KAT-Coder-V2.5-Dev-GGUFOpen weightsbartowskiestimated: 34.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.7Bfp1687.7 GB+102.3 GB✓ fits
perplexity-ai/pplx-computer-qwen-3-6-35b-a3b-mlx-20260709Open weightsPerplexity AIestimated: 34.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.7Bfp1687.7 GB+102.3 GB✓ fits
Nous-Hermes-2-Yi-34BOpen weightsNous Researchestimated: 34.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.4Bfp1687.1 GB+102.9 GB✓ fits
GKA-primed-HQwen3-32B-ReasonerOpen weightsAmazon Web Servicesestimated: 34.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens34.1Bfp1686.5 GB+103.5 GB✓ fits
MiniMax-H3Open weightsMiniMaxestimated: 33.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens33.1Bfp1684.2 GB+105.8 GB✓ fits
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8Open weightsNVIDIAestimated: 33.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens33Bfp1683.9 GB+106.1 GB✓ fits
SynLogic-Mix-3-32BOpen weightsMiniMaxestimated: 32.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.8Bfp1683.4 GB+106.6 GB✓ fits
DeepSeek-R1-Distill-Qwen-32BOpen weightsDeepSeekestimated: 32.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.8Bfp1683.4 GB+106.6 GB✓ fits
SynLogic-32BOpen weightsMiniMaxestimated: 32.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.8Bfp1683.4 GB+106.6 GB✓ fits
Qwen3 32BOpen weightsQwenestimated: 32.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.8Bfp1683.3 GB+106.7 GB✓ fits
c4ai-command-r-08-2024RestrictedCohereestimated: 32.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.3Bfp1682.3 GB+107.7 GB✓ fits
aya-expanse-32bRestrictedCohereestimated: 32.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.3Bfp1682.3 GB+107.7 GB✓ fits
OLMo-2-0325-32B-InstructOpen weightsAllen Institute for AIestimated: 32.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.2Bfp1682.1 GB+107.9 GB✓ fits
FLUX.2-devRestrictedBlack Forest Labsestimated: 32.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens32.2Bfp1682.1 GB+107.9 GB✓ fits
Nemotron 3 Nano 30B A3BOpen weightsNVIDIAestimated: 31.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens31.6Bfp1680.6 GB+109.4 GB✓ fits
Gemma 4 31BOpen weightsGoogleestimated: 31.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens31.3Bfp1679.9 GB+110.1 GB✓ fits
mlx-community/gemma-4-31b-it-4bitOpen weightsMLX Communityestimated: 31.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens31.3Bfp1679.9 GB+110.1 GB✓ fits
GLM 4.7 FlashOpen weightsZ.ai (Zhipu AI)estimated: 31.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens31.2Bfp1679.8 GB+110.2 GB✓ fits
unsloth/gemma-4-31B-it-qat-GGUFOpen weightsUnslothestimated: 30.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.7Bfp1678.6 GB+111.4 GB✓ fits
mlx-community/Qwen3-30B-A3B-Instruct-2507-4bitOpen weightsMLX Communityestimated: 30.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.5Bfp1678.2 GB+111.8 GB✓ fits
unsloth/Qwen3-30B-A3B-Thinking-2507-GGUFOpen weightsUnslothestimated: 30.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.5Bfp1678.2 GB+111.8 GB✓ fits
browsesafeOpen weightsPerplexity AIestimated: 30.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.5Bfp1678.2 GB+111.8 GB✓ fits
unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUFOpen weightsUnslothestimated: 30.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.5Bfp1678.2 GB+111.8 GB✓ fits
North Mini Code (free)Open weightsCohereestimated: 30.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.5Bfp1678.1 GB+111.9 GB✓ fits
Hy-MT2-30B-A3BOpen weightsTencentestimated: 30.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens30.1Bfp1677.2 GB+112.8 GB✓ fits
ibm-granite/granite-4.2-30b-GGUFOpen weightsIBMestimated: 30.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens30Bfp1677.0 GB+113.0 GB✓ fits
north-mini-code-1-0Cohereestimated: 30.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens30Bfp1677.0 GB+113.0 GB✓ fits
ERNIE-4.5-VL-28B-A3B-ThinkingOpen weightsBaiduestimated: 29.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens29.7Bfp1676.2 GB+113.8 GB✓ fits
ERNIE-4.5-VL-28B-A3B-PTOpen weightsBaiduestimated: 29.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens29.4Bfp1675.6 GB+114.4 GB✓ fits
granite-4.1-30bOpen weightsIBMestimated: 28.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens28.9Bfp1674.4 GB+115.6 GB✓ fits
Qwen/Qwen3.6-27B-FP8Open weightsQwenestimated: 27.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.8Bfp1671.9 GB+118.1 GB✓ fits
Qwen/Qwen3.8-27B-FP8Open weightsQwenestimated: 27.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.8Bfp1671.9 GB+118.1 GB✓ fits
Qwen3.8 27BOpen weightsQwenestimated: 27.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.8Bfp1671.9 GB+118.1 GB✓ fits
Qwen3.6 27BOpen weightsQwenestimated: 27.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.8Bfp1671.9 GB+118.1 GB✓ fits
mlx-community/Qwen3.8-27B-4bitOpen weightsMLX Communityestimated: 27.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.4Bfp1670.9 GB+119.1 GB✓ fits
mlx-community/Qwen3.8-27B-8bitOpen weightsMLX Communityestimated: 27.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.4Bfp1670.9 GB+119.1 GB✓ fits
unsloth/Qwen3.6-27B-MTP-GGUFOpen weightsUnslothestimated: 27.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.3Bfp1670.8 GB+119.2 GB✓ fits
unsloth/Qwen3.8-27B-GGUFOpen weightsUnslothestimated: 27.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.3Bfp1670.8 GB+119.2 GB✓ fits
perplexity-ai/pplx-computer-qwen-3-8-27b-dflash2-gguf-20260826Open weightsPerplexity AIestimated: 27.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens27.3Bfp1670.8 GB+119.2 GB✓ fits
bartowski/Qwen3.8-27B-GGUFOpen weightsbartowskiestimated: 27.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens27Bfp1670.1 GB+119.9 GB✓ fits
bartowski/Fara1.5-27B-GGUFOpen weightsbartowskiestimated: 27.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens27Bfp1670.1 GB+119.9 GB✓ fits
unsloth/Qwen3.6-27B-GGUFOpen weightsUnslothestimated: 26.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens26.9Bfp1669.9 GB+120.1 GB✓ fits
Gemma 4 26B A4BOpen weightsGoogleestimated: 25.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens25.8Bfp1667.3 GB+122.7 GB✓ fits
unsloth/gemma-4-26B-A4B-it-GGUFOpen weightsUnslothestimated: 25.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens25.2Bfp1666.0 GB+124.0 GB✓ fits
unsloth/gemma-4-26B-A4B-it-qat-GGUFOpen weightsUnslothestimated: 25.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens25.2Bfp1666.0 GB+124.0 GB✓ fits
cerebras/Qwen3-Coder-REAP-25B-A3BOpen weightsCerebras Systemsestimated: 24.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens24.9Bfp1665.2 GB+124.8 GB✓ fits
unsloth/Qwen3.6-35B-A3B-NVFP4Open weightsUnslothestimated: 24.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens24.6Bfp1664.7 GB+125.3 GB✓ fits
Voxtral Small 24B 2507Open weightsMistral AIestimated: 24.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens24.3Bfp1663.8 GB+126.2 GB✓ fits
Devstral-Small-2-24B-Instruct-2512Open weightsMistral AIestimated: 24.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens24Bfp1663.2 GB+126.8 GB✓ fits
mlx-community/Devstral-Small-2-24B-Instruct-2512-4bitOpen weightsMLX Communityestimated: 24.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens24Bfp1663.2 GB+126.8 GB✓ fits
Mistral Small 3.2 24BOpen weightsMistral AIestimated: 24.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens24Bfp1663.2 GB+126.8 GB✓ fits
Mistral Small 3.1 24BOpen weightsMistral AIestimated: 24.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens24Bfp1663.2 GB+126.8 GB✓ fits
cerebras/GLM-4.7-Flash-REAP-23B-A3BOpen weightsCerebras Systemsestimated: 23.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens23Bfp1660.9 GB+129.1 GB✓ fits
ERNIE-4.5-21B-A3B-PTOpen weightsBaiduestimated: 21.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens21.9Bfp1658.5 GB+131.5 GB✓ fits
ERNIE-4.5-21B-A3B-Base-PTOpen weightsBaiduestimated: 21.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens21.8Bfp1658.2 GB+131.8 GB✓ fits
ERNIE-4.5-21B-A3B-ThinkingOpen weightsBaiduestimated: 21.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens21.8Bfp1658.2 GB+131.8 GB✓ fits
gpt-oss-safeguard-20bOpen weightsOpenAIestimated: 21.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens21.5Bfp1657.5 GB+132.5 GB✓ fits
unsloth/Qwen3.6-27B-NVFP4Open weightsUnslothestimated: 21.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens21.2Bfp1656.8 GB+133.2 GB✓ fits
gpt-oss-20bOpen weightsOpenAIestimated: 20.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens20.9Bfp1656.1 GB+133.9 GB✓ fits
mlx-community/gpt-oss-20b-MXFP4-Q8Open weightsMLX Communityestimated: 20.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens20.9Bfp1656.1 GB+133.9 GB✓ fits
nvidia/Gemma-4-31B-IT-NVFP4Open weightsNVIDIAestimated: 20.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens20.9Bfp1656.0 GB+134.0 GB✓ fits
gpt-neox-20bOpen weightsEleutherAIestimated: 20.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens20.7Bfp1655.7 GB+134.3 GB✓ fits
unsloth/MiniMax-H3-GGUFOpen weightsUnslothestimated: 20.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens20.1Bfp1654.3 GB+135.7 GB✓ fits
internlm2-20bOpen weightsInternLM (Shanghai AI Laboratory)estimated: 20.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens20Bfp1654.0 GB+136.0 GB✓ fits
internlm2-base-20bOpen weightsInternLM (Shanghai AI Laboratory)estimated: 20.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens20Bfp1654.0 GB+136.0 GB✓ fits
internlm/internlm2_5-20b-chat-ggufOpen weightsInternLM (Shanghai AI Laboratory)estimated: 20.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens20Bfp1654.0 GB+136.0 GB✓ fits
unsloth/Qwen3.8-27B-NVFP4Open weightsUnslothestimated: 19.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens19.9Bfp1653.7 GB+136.3 GB✓ fits
internlm2-chat-20bOpen weightsInternLM (Shanghai AI Laboratory)estimated: 19.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens19.9Bfp1653.7 GB+136.3 GB✓ fits
Intel/Qwen3.5-122B-A10B-int4-AutoRoundOpen weightsIntelestimated: 19.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens19.7Bfp1653.2 GB+136.8 GB✓ fits
pplx-qwen-3-8-27b-dflash2-20260819Open weightsPerplexity AIestimated: 18.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens18.8Bfp1651.2 GB+138.8 GB✓ fits
pplx-computer-qwen-3-8-27b-dflash2-20260824Open weightsPerplexity AIestimated: 18.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens18.8Bfp1651.2 GB+138.8 GB✓ fits
nvidia/Qwen3.6-35B-A3B-NVFP4Open weightsNVIDIAestimated: 18.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens18.7Bfp1651.0 GB+139.0 GB✓ fits
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4Open weightsNVIDIAestimated: 18.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens18.3Bfp1650.1 GB+139.8 GB✓ fits
nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4Open weightsNVIDIAestimated: 17.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens17.8Bfp1649.0 GB+141.0 GB✓ fits
Kimi-VL-A3B-ThinkingOpen weightsMoonshot AIestimated: 16.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens16.4Bfp1645.7 GB+144.3 GB✓ fits
Kimi-VL-A3B-InstructOpen weightsMoonshot AIestimated: 16.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens16.4Bfp1645.7 GB+144.3 GB✓ fits
Kimi-VL-A3B-Thinking-2506Open weightsMoonshot AIestimated: 16.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens16.4Bfp1645.7 GB+144.3 GB✓ fits
Moonlight-16B-A3B-InstructOpen weightsMoonshot AIestimated: 16.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens16Bfp1644.7 GB+145.3 GB✓ fits
Moonlight-16B-A3BOpen weightsMoonshot AIestimated: 16.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens16Bfp1644.7 GB+145.3 GB✓ fits
DeepSeek-Coder-V2-Lite-InstructOpen weightsDeepSeekestimated: 15.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens15.7Bfp1644.1 GB+145.9 GB✓ fits
DeepSeek-V2-LiteOpen weightsDeepSeekestimated: 15.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens15.7Bfp1644.1 GB+145.9 GB✓ fits
bartowski/DeepSeek-Coder-V2-Lite-Instruct-GGUFOpen weightsbartowskiestimated: 15.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens15.7Bfp1644.1 GB+145.9 GB✓ fits
amd/Qwen3.8-27B-Quark-AWQ-MXFP4Open weightsAMDestimated: 15.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens15.6Bfp1643.9 GB+146.1 GB✓ fits
BAGEL-7B-MoTOpen weightsByteDanceestimated: 14.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens14.7Bfp1641.8 GB+148.2 GB✓ fits
Phi 4Open weightsMicrosoftestimated: 14.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens14.7Bfp1641.7 GB+148.3 GB✓ fits
nvidia/Gemma-4-26B-A4B-NVFP4Open weightsNVIDIAestimated: 14.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens14.4Bfp1641.1 GB+148.9 GB✓ fits
Ministral 3 14B 2512Open weightsMistral AIestimated: 13.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens13.9Bfp1640.1 GB+149.9 GB✓ fits
Ministral-3-14B-Reasoning-2512Open weightsMistral AIestimated: 13.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens13.9Bfp1640.1 GB+149.9 GB✓ fits
FireLLaVA-13bRestrictedFireworks AIestimated: 13.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens13.3Bfp1638.7 GB+151.3 GB✓ fits
gemma-4-12B-it-qat-w4a16-ctOpen weightsGoogleestimated: 13.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens13.3Bfp1638.6 GB+151.4 GB✓ fits
Llama-2-13b-chat-hfRestrictedMeta AIestimated: 13.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens13Bfp1637.9 GB+152.1 GB✓ fits
aya-101Open weightsCohereestimated: 12.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens12.9Bfp1637.7 GB+152.3 GB✓ fits
Mistral NemoOpen weightsMistral AIestimated: 12.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens12.3Bfp1636.2 GB+153.8 GB✓ fits
Mistral-Nemo-Base-2407Open weightsMistral AIestimated: 12.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens12.3Bfp1636.2 GB+153.8 GB✓ fits
mlx-community/gemma-3-12b-it-4bitOpen weightsMLX Communityestimated: 12.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens12Bfp1635.6 GB+154.4 GB✓ fits
bartowski/kai-os_Grug-12B-GGUFOpen weightsbartowskiestimated: 12.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens12Bfp1635.6 GB+154.4 GB✓ fits
gemma-4-12B-itOpen weightsGoogleestimated: 12.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens12Bfp1635.5 GB+154.5 GB✓ fits
unsloth/gemma-4-12B-it-qat-GGUFOpen weightsUnslothestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.6 GB✓ fits
FLUX.1-Fill-devRestrictedBlack Forest Labsestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.6 GB✓ fits
FLUX.1-devRestrictedBlack Forest Labsestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.6 GB✓ fits
FLUX.1-Krea-devRestrictedBlack Forest Labsestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.6 GB✓ fits
FLUX.1-Kontext-devRestrictedBlack Forest Labsestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.6 GB✓ fits
FLUX.1-schnellRestrictedBlack Forest Labsestimated: 11.9B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.9Bfp1635.4 GB+154.7 GB✓ fits
Intel/Qwen3-Coder-Next-int4-AutoRoundOpen weightsIntelestimated: 11.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.8Bfp1635.2 GB+154.8 GB✓ fits
KaLM-Embedding-Gemma3-12B-2511Open weightsTencentestimated: 11.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens11.8Bfp1635.1 GB+154.9 GB✓ fits
Nous-Hermes-2-SOLAR-10.7BOpen weightsNous Researchestimated: 10.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens10.7Bfp1632.7 GB+157.3 GB✓ fits
Llama-3.2-11B-Vision-InstructRestrictedMeta AIestimated: 10.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens10.7Bfp1632.5 GB+157.5 GB✓ fits
GLM-4.1V-9B-ThinkingOpen weightsZ.ai (Zhipu AI)estimated: 10.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens10.3Bfp1631.7 GB+158.3 GB✓ fits
GLM-4.6V-FlashOpen weightsZ.ai (Zhipu AI)estimated: 10.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens10.3Bfp1631.7 GB+158.3 GB✓ fits
Kimi-Audio-7B-InstructOpen weightsMoonshot AIestimated: 9.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.77Bfp1630.5 GB+159.5 GB✓ fits
Kimi-Audio-7BOpen weightsMoonshot AIestimated: 9.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.77Bfp1630.5 GB+159.5 GB✓ fits
Qwen3.5-9BOpen weightsQwenestimated: 9.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.65Bfp1630.2 GB+159.8 GB✓ fits
glm-4-9b-chatOpen weightsZ.ai (Zhipu AI)estimated: 9.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.4Bfp1629.6 GB+160.4 GB✓ fits
academic-ds-9BOpen weightsByteDanceestimated: 9.4B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.37Bfp1629.5 GB+160.5 GB✓ fits
FLUX.2-klein-9BRestrictedBlack Forest Labsestimated: 9.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.08Bfp1628.9 GB+161.1 GB✓ fits
FLUX.2-klein-base-9BRestrictedBlack Forest Labsestimated: 9.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.08Bfp1628.9 GB+161.1 GB✓ fits
FLUX.2-klein-9b-kvRestrictedBlack Forest Labsestimated: 9.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens9.08Bfp1628.9 GB+161.1 GB✓ fits
black-forest-labs/FLUX.2-klein-9b-kv-fp8Open weightsBlack Forest Labsestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens9Bfp1628.7 GB+161.3 GB✓ fits
black-forest-labs/FLUX.2-klein-9b-fp8RestrictedBlack Forest Labsestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens9Bfp1628.7 GB+161.3 GB✓ fits
bartowski/Fara1.5-9B-GGUFOpen weightsbartowskiestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens9Bfp1628.7 GB+161.3 GB✓ fits
black-forest-labs/FLUX.2-klein-base-9b-fp8RestrictedBlack Forest Labsestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens9Bfp1628.7 GB+161.3 GB✓ fits
bartowski/tencent_UI-Mate-9B-GGUFOpen weightsbartowskiestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens9Bfp1628.7 GB+161.3 GB✓ fits
unsloth/Qwen3.5-9B-GGUFOpen weightsUnslothestimated: 9.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.95Bfp1628.6 GB+161.4 GB✓ fits
Qianfan-VL-8BOpen weightsBaiduestimated: 8.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.81Bfp1628.3 GB+161.7 GB✓ fits
internlm3-8b-instructOpen weightsInternLM (Shanghai AI Laboratory)estimated: 8.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.8Bfp1628.3 GB+161.8 GB✓ fits
granite-4.1-8bOpen weightsIBMestimated: 8.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.79Bfp1628.2 GB+161.8 GB✓ fits
Qwen3 VL 8B InstructOpen weightsQwenestimated: 8.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.77Bfp1628.2 GB+161.8 GB✓ fits
amd/Qwen3-VL-8B-Instruct-w8a8-llmcompressorOpen weightsAMDestimated: 8.8B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.77Bfp1628.2 GB+161.8 GB✓ fits
VibeVoice-ASROpen weightsMicrosoftestimated: 8.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.67Bfp1627.9 GB+162.1 GB✓ fits
Molmo2-8BOpen weightsAllen Institute for AIestimated: 8.7B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.66Bfp1627.9 GB+162.1 GB✓ fits
aya-vision-8bRestrictedCohereestimated: 8.6B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.63Bfp1627.9 GB+162.2 GB✓ fits
Intern-S1-miniOpen weightsInternLM (Shanghai AI Laboratory)estimated: 8.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.54Bfp1627.6 GB+162.4 GB✓ fits
GKA-primed-HQwen3-8B-ReasonerOpen weightsAmazon Web Servicesestimated: 8.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.5Bfp1627.6 GB+162.4 GB✓ fits
GDN-primed-HQwen3-8B-InstructOpen weightsAmazon Web Servicesestimated: 8.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.5Bfp1627.6 GB+162.4 GB✓ fits
LFM2.5-8B-A1BOpen weightsLiquid AIestimated: 8.5B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.47Bfp1627.5 GB+162.5 GB✓ fits
allenai/olmOCR-2-7B-1025-FP8Open weightsAllen Institute for AIestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
Qwen2.5-VL-7B-InstructOpen weightsQwenestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
UI-TARS 7BOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
olmOCR-2-7B-1025Open weightsAllen Institute for AIestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
UI-TARS-7B-SFTOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
UI-TARS-7B-DPOOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.29Bfp1627.1 GB+162.9 GB✓ fits
Seed-Coder-8B-BaseOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.25Bfp1627.0 GB+163.0 GB✓ fits
Stable-DiffCoder-8B-InstructOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.25Bfp1627.0 GB+163.0 GB✓ fits
Seed-Coder-8B-InstructOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.25Bfp1627.0 GB+163.0 GB✓ fits
Stable-DiffCoder-8B-BaseOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.25Bfp1627.0 GB+163.0 GB✓ fits
Seed-Coder-8B-ReasoningOpen weightsByteDanceestimated: 8.3B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.25Bfp1627.0 GB+163.0 GB✓ fits
DeepSeek-R1-0528-Qwen3-8BOpen weightsDeepSeekestimated: 8.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.19Bfp1626.8 GB+163.2 GB✓ fits
Qwen3 8BOpen weightsQwenestimated: 8.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.19Bfp1626.8 GB+163.2 GB✓ fits
mlx-community/Qwen3-8B-4bitOpen weightsMLX Communityestimated: 8.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.19Bfp1626.8 GB+163.2 GB✓ fits
granite-guardian-3.3-8bOpen weightsIBMestimated: 8.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.17Bfp1626.8 GB+163.2 GB✓ fits
granite-3.0-8b-instructOpen weightsIBMestimated: 8.2B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.17Bfp1626.8 GB+163.2 GB✓ fits
stable-diffusion-3.5-largeRestrictedStability AIestimated: 8.1B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.15Bfp1626.7 GB+163.3 GB✓ fits
ERNIE-ImageOpen weightsBaiduestimated: 8.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.03Bfp1626.5 GB+163.5 GB✓ fits
ERNIE-Image-TurboOpen weightsBaiduestimated: 8.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.03Bfp1626.5 GB+163.5 GB✓ fits
NousResearch/Hermes-3-Llama-3.1-8B-GGUFOpen weightsNous Researchestimated: 8.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.03Bfp1626.5 GB+163.5 GB✓ fits
NousResearch/Meta-Llama-3-8B-InstructOpen weightsNous Researchestimated: 8.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.03Bfp1626.5 GB+163.5 GB✓ fits
NousResearch/Meta-Llama-3.1-8BOpen weightsNous Researchestimated: 8.0B params × 2.0 B × 1.15 + KV cache for 131072 tokens8.03Bfp1626.5 GB+163.5 GB✓ fits

Headroom = device memory − 2 GB reserve − estimated need. Sorted by the API (fitting models first). Compare the models you shortlist with the + buttons. Devices with these memory sizes: hardware listing.