What can you run with 64GB of RAM?
With 64GB RAM you can run 94 of the 129 open-source models we track — up to about 91B parameters at Q4 quantization. The most popular that fits is Qwen3.5-9B, needing 6.8 GB and generating around ~10 tok/s on typical hardware of this size.
Typical machines: MacBook Pro M4 Max · Mac Studio · High-end desktop. After the operating system takes its share, about 64 GB is free for a model.
Having the memory is not the same as having the speed
64GB RAM decides whether a model loads. How fast it answers is decided by memory bandwidth — how quickly your machine can read the model, which it must do once for every word it writes. Machines with 64GB RAM span 90–614 GB/s, so the fastest is about 6.8× quicker than the slowest with identical capacity. The table quotes a DDR5-5600; pick yours below to see the difference. Why bandwidth and not the processor →
90 GB/sDesktop or laptop — DDR5-5600· 5,600 MT/s × 128-bit dual channel ÷ 8 = 89.6 GB/s
128 GB/sEnthusiast desktop — DDR5-8000· 8,000 MT/s × 128-bit dual channel ÷ 8 = 128 GB/s
150 GB/sMac — M3 Pro· Apple published specification for the M3 Pro: 150 GB/s (lower than the M2 Pro)
200 GB/sMac — M1 Pro· Apple published specification for the M1 Pro: 200 GB/s
200 GB/sMac — M2 Pro· Apple published specification for the M2 Pro: 200 GB/s
273 GB/sMac — M4 Pro· Apple published specification for the M4 Pro: 273 GB/s
307 GB/sMac — M5 Pro· Apple published specification for the M5 Pro: up to 307 GB/s
400 GB/sMac — M1 Max· Apple published specification for the M1 Max: 400 GB/s
400 GB/sMac — M2 Max· Apple published specification for the M2 Max: 400 GB/s
400 GB/sMac — M3 Max· Apple published specification for the M3 Max: 400 GB/s (16-core GPU; the 14-core is 300 GB/s)
546 GB/sMac — M4 Max· Apple published specification for the M4 Max: 546 GB/s
614 GB/sMac — M5 Max· Apple published specification for the M5 Max: 614 GB/s (40-core GPU; the 32-core is 460 GB/s)
Models that run on 64GB RAM in 2026
| Model↕ | Size↕ | Memory↑ | Est. speed↕ | Downloads↕ | Feels like |
|---|---|---|---|---|---|
| granite-embedding-97m-multilingual-r2Ibm | 97M | 603 MBRuns well | ~994 tok/s | 97,079 | Fast |
| LFM2.5-Encoder-230MLiquid Ai | 230M | 793 MBRuns well | ~422 tok/s | 13,432 | Fast |
| LFM2.5-230MLiquid Ai | 230M | 808 MBRuns well | ~381 tok/s | 55,999 | Fast |
| granite-embedding-311m-multilingual-r2Ibm | 312M | 821 MBRuns well | ~231 tok/s | 64,530 | Fast |
| granite-speech-5.0-470m-turboctcIbm | 473M | 857 MBRuns well | ~205 tok/s | 11,294 | Fast |
| harrier-oss-v1-270mMicrosoft | 268M | 865 MBRuns well | ~231 tok/s | 727,222 | Fast |
| LFM2.5-Encoder-350M-Policy-LinterLiquid Ai | 355M | 885 MBRuns well | ~273 tok/s | 9,957 | Fast |
| LFM2.5-VL-450MLiquid Ai | 449M | 900 MBRuns well | ~255 tok/s | 67,396 | Fast |
| nemotron-3.5-asr-streaming-0.6bNVIDIA | 638M | 1.1 GBRuns well | ~118 tok/s | 917,259 | Fast |
| Qwen3.5-0.8BAlibaba / Qwen | 873M | 1.3 GBRuns well | ~110 tok/s | 2,881,964 | Fast |
| privacy-filterOpenAI · MoE | 1.4B | 1.4 GBRuns well | ~346 tok/s280M active | 456,933 | Fast |
| LFM2.5-1.2B-InstructLiquid Ai | 1.2B | 1.4 GBRuns well | ~80 tok/s | 366,847 | Fast |
| LFM2.5-VL-1.6BLiquid Ai | 1.6B | 1.4 GBRuns well | ~80 tok/s | 248,110 | Fast |
| Nemotron-3-Embed-1B-BF16NVIDIA | 1.1B | 1.5 GBRuns well | ~85 tok/s | 386,914 | Fast |
| Hy-MT2-1.8BTencent | 2B | 1.9 GBRuns well | ~52 tok/s | 24,475 | Fast |
| MiniMax-Music3MiniMax | 2.4B | 2.1 GBRuns well | ~42 tok/s | 23,217 | Fast |
| Qwen3.5-2BAlibaba / Qwen | 2.3B | 2.1 GBRuns well | ~42 tok/s | 3,097,484 | Fast |
| cohere-transcribe-arabic-07-2026Cohere | 2.1B | 2.3 GBRuns well | ~38 tok/s | 51,893 | Fast |
| cohere-transcribe-03-2026Cohere | 2.1B | 2.3 GBRuns well | ~38 tok/s | 607,964 | Fast |
| granite-speech-4.1-2bIbm | 2.3B | 2.4 GBRuns well | ~37 tok/s | 237,181 | Fast |
| LFM2.5-2.6BLiquid Ai | 2.7B | 2.5 GBRuns well | ~35 tok/s | 133,677 | Fast |
| LFM2.5-VL-3BLiquid Ai | 3.1B | 2.5 GBRuns well | ~35 tok/s | 24,282 | Fast |
| granite-4.0-1b-speechIbm | 2.3B | 2.5 GBRuns well | ~36 tok/s | 74,644 | Fast |
| LFM2.5-Audio-1.5BLiquid Ai | 1.5B | 2.6 GBRuns well | ~32 tok/s | 1,702 | Fast |
| Unlimited-OCRBaidu AI | 3.3B | 2.7 GBRuns well | ~30 tok/s | 3,008,635 | Fast |
| Nemotron-Labs-Diffusion-3BNVIDIA | 3B | 2.8 GBRuns well | ~32 tok/s | 35,414 | Fast |
| LocateAnything-3BNVIDIA | 3.8B | 2.8 GBRuns well | ~28 tok/s | 95,054 | Fast |
| granite-4.1-3bIbm | 3.4B | 2.9 GBRuns well | ~28 tok/s | 122,076 | Fast |
| granite-4.2-3bIbm | 3.7B | 3.1 GBRuns well | ~26 tok/s | 14,073 | Fast |
| granite-4.0-3b-visionIbm | 4B | 3.1 GBRuns well | ~26 tok/s | 1,069 | Fast |
| Shieldstral-1.0-3BMistral AI | 3.8B | 3.1 GBRuns well | ~27 tok/s | 20,895 | Fast |
| Ministral-3-3B-Instruct-2512Mistral AI | 3.8B | 3.1 GBRuns well | ~27 tok/s | 544,419 | Fast |
| Nemotron-Labs-Diffusion-3B-BaseNVIDIA | 3.8B | 3.3 GBRuns well | ~25 tok/s | 19,213 | Fast |
| Cosmos3-EdgeNVIDIA | 3.9B | 3.3 GBRuns well | ~25 tok/s | 1,325,763 | Fast |
| Nemotron-3.5-Content-SafetyNVIDIA | 4.3B | 3.6 GBRuns well | ~23 tok/s | 134,547 | Comfortable |
| gemma-4-E2B-itGoogle DeepMind | 5.1B | 3.8 GBRuns well | ~19 tok/s | 3,235,641 | Comfortable |
| Qwen3.5-4BAlibaba / Qwen | 4.7B | 3.8 GBRuns well | ~21 tok/s | 7,681,585 | Comfortable |
| Fara1.5-4BMicrosoft | 4.5B | 4.0 GBRuns well | ~20 tok/s | 2,786 | Comfortable |
| MolmoWeb-4BAi2 | 4.9B | 4.1 GBRuns well | ~20 tok/s | 2,152 | Comfortable |
| NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA | 4B | 4.1 GBRuns well | ~21 tok/s | 712,701 | Comfortable |
| Fara-7BMicrosoft | 8.3B | 5.5 GBRuns well | ~12 tok/s | 1,593 | Comfortable |
| Hy-MT2-7BTencent | 8B | 5.7 GBRuns well | ~13 tok/s | 12,004 | Comfortable |
| LFM2.5-8B-A1BLiquid Ai · MoE | 8.5B | 5.9 GBRuns well | ~96 tok/s1B active | 96,675 | Fast |
| Nemotron-3-Embed-8B-BF16NVIDIA | 8B | 5.9 GBRuns well | ~12 tok/s | 81,804 | Comfortable |
| Nemotron-Labs-Diffusion-8BNVIDIA | 8B | 5.9 GBRuns well | ~12 tok/s | 163,233 | Comfortable |
| Cosmos3-NanoNVIDIA | 16B | 6.2 GBRuns well | ~12 tok/s | 244,209 | Comfortable |
| gemma-4-E4B-itGoogle DeepMind | 8B | 6.2 GBRuns well | ~11 tok/s | 4,740,694 | Comfortable |
| Nemotron-Labs-Diffusion-8B-BaseNVIDIA | 8.5B | 6.2 GBRuns well | ~11 tok/s | 280,417 | Comfortable |
| Ministral-3-8B-Instruct-2512Mistral AI | 8.9B | 6.3 GBRuns well | ~11 tok/s | 119,379 | Comfortable |
| granite-4.1-8bIbm | 8.8B | 6.3 GBRuns well | ~11 tok/s | 1,441,995 | Comfortable |
| MolmoWeb-8BAi2 | 8.7B | 6.4 GBRuns well | ~11 tok/s | 1,679 | Comfortable |
| granite-4.2-8bIbm | 8.8B | 6.6 GBRuns well | ~11 tok/s | 12,050 | Comfortable |
| Ornith-1.0-9BDeepreinforce | 9B | 6.7 GBRuns well | ~10 tok/s | 2,250,422 | Comfortable |
| Qwen3.5-9BAlibaba / Qwen | 9.7B | 6.8 GBRuns well | ~10 tok/s | 13,660,443 | Comfortable |
| GLM-4.6V-FlashZhipu AI | 10B | 7.0 GBRuns well | ~9.5 tok/s | 282,562 | Comfortable |
| Fara1.5-9BMicrosoft | 9.4B | 7.0 GBRuns well | ~9.9 tok/s | 4,461 | Comfortable |
| Olmo-3-7B-InstructAi2 | 7.3B | 7.2 GBRuns well | ~13 tok/s | 406,776 | Comfortable |
| gemma-4-12B-itGoogle DeepMind | 12B | 9.6 GBRuns well | ~8.1 tok/s | 3,165,860 | Comfortable |
| Ministral-3-14B-Instruct-2512Mistral AI | 14B | 9.7 GBRuns well | ~7.1 tok/s | 198,169 | Slow |
| Nemotron-Labs-Diffusion-14BNVIDIA | 14B | 10.0 GBRuns well | ~6.9 tok/s | 12,153 | Slow |
| NVIDIA-NemotronLabs-VoiceChat-11BNVIDIA | 11B | 15 GBRuns well | ~4.9 tok/s | 3,459 | Slow |
| Voxtral-Small-24B-2507Mistral AI | 24B | 16 GBRuns well | ~4.1 tok/s | 200,361 | Slow |
| Qwen3.8-27BAlibaba / Qwen | 28B | 19 GBRuns well | ~3.6 tok/s | 5,254,882 | Slow |
| Qwen3.5-27BAlibaba / Qwen | 28B | 19 GBRuns well | ~3.5 tok/s | 2,817,727 | Slow |
| diffusiongemma-26B-A4B-itGoogle DeepMind · MoE | 26B | 19 GBRuns well | ~22 tok/s4B active | 1,170,594 | Comfortable |
| gemma-4-26B-A4B-itGoogle DeepMind · MoE | 26B | 20 GBRuns well | ~22 tok/s4B active | 7,837,690 | Comfortable |
| granite-4.1-30bIbm | 29B | 20 GBRuns well | ~3.3 tok/s | 177,103 | Slow |
| Fara1.5-27BMicrosoft | 27B | 20 GBRuns well | ~3.3 tok/s | 2,458 | Slow |
| Hy-MT2-30B-A3BTencent · MoE | 30B | 20 GBRuns well | ~32 tok/s3B active | 22,910 | Fast |
| granite-4.2-30bIbm | 29B | 21 GBRuns well | ~3.3 tok/s | 5,244 | Slow |
| North-Mini-Code-1.0Cohere · MoE | 30B | 22 GBRuns well | ~15 tok/s6.1B active | 11,421 | Comfortable |
| Gemma-4-26B-A4B-NVFP4NVIDIA · MoE | 14B | 22 GBRuns well | ~11 tok/s4B active | 1,560,310 | Comfortable |
| diffusiongemma-26B-A4B-it-NVFP4NVIDIA · MoE | 14B | 22 GBRuns well | ~11 tok/s4B active | 108,924 | Comfortable |
| GLM-4.7-FlashZhipu AI | 31B | 22 GBRuns well | ~3.1 tok/s | 2,071,716 | Slow |
| Olmo-3.1-32B-InstructAi2 | 32B | 23 GBRuns well | ~3.0 tok/s | 51,153 | Slow |
| Ornith-1.0-35BDeepreinforce · MoE | 35B | 24 GBRuns well | ~14 tok/s7B active | 2,736,644 | Comfortable |
| Gemma-4-31B-IT-NVFP4NVIDIA | 21B | 24 GBRuns well | ~3.3 tok/s | 2,388,034 | Slow |
| Qwen3.6-35B-A3BAlibaba / Qwen · MoE | 36B | 25 GBRuns well | ~32 tok/s3B active | 5,623,400 | Fast |
| Qwen-AgentWorld-35B-A3BAlibaba / Qwen · MoE | 35B | 25 GBRuns well | ~31 tok/s3B active | 60,898 | Fast |
| gemma-4-31B-itGoogle DeepMind | 31B | 25 GBRuns well | ~3.1 tok/s | 7,973,406 | Slow |
| Qwen3.5-35B-A3BAlibaba / Qwen · MoE | 36B | 25 GBRuns well | ~31 tok/s3B active | 2,437,252 | Fast |
| Alpamayo2-SuperNVIDIA | 36B | 26 GBRuns well | ~2.7 tok/s | 17,622 | Very slow |
| Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4NVIDIA · MoE | 18B | 27 GBRuns well | ~16 tok/s3B active | 1,570,072 | Comfortable |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4NVIDIA · MoE | 18B | 28 GBRuns well | ~14 tok/s3B active | 1,064,894 | Comfortable |
| North-Mini-Code-1.0-fp8Cohere · MoE | 31B | 36 GBRuns well | ~9.1 tok/s6.1B active | 35,994 | Comfortable |
| Qwen3.6-27BAlibaba / Qwen | 28B | 41 GBRuns well | ~1.6 tok/s | 6,615,659 | Very slow |
| Cosmos3-Super-Text2Image-4StepNVIDIA | 64B | 43 GBRuns well | ~1.5 tok/s | 97,839 | Very slow |
| Cosmos3-Super-Image2VideoNVIDIA | 65B | 43 GBRuns well | ~1.5 tok/s | 75,852 | Very slow |
| Cosmos3-SuperNVIDIA | 65B | 43 GBRuns well | ~1.5 tok/s | 117,430 | Very slow |
| Nemotron-Labs-TwoTower-30B-A3B-Base-BF16NVIDIA · MoE | 63B | 46 GBRuns well | ~31 tok/s3B active | 793 | Fast |
| Qwen3-Next-80B-A3B-InstructAlibaba / Qwen · MoE | 81B | 52 GBRuns well | ~24 tok/s3B active | 287,898 | Comfortable |
| gpt-oss-puzzle-88BNVIDIA | 91B | 54 GBRuns well | ~1.2 tok/s | 140,544 | Very slow |
| NVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-NVFP4NVIDIA · MoE | 45B | 63 GBTight | ~5.4 tok/s9B active | 345,084 | Slow |
| MiniMax-H3MiniMax | 33B | 64 GBTight | ~1.1 tok/s | 5,092,067 | Very slow |
Sorted by memory, smallest first — click any column to change it.
Upgrade
Step up to 128GB RAM and 7 more models come within reach
Including Qwen3.5-122B-A10B (122B), NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 (67B) and Step-3.5-Flash (199B) — around ~59 tok/s on a typical machine that size.
See everything 128GB RAM runs →All figures at Q4_K_M quantization, 4K context. “Memory” includes the model, its context cache and runtime overhead. Speed is calculated, not benchmarked: 90 GB/s × 0.65 ÷ bytes read per token, assuming a DDR5-5600. Scale it by your own machine's bandwidth from the list above. Mixture-of-experts models read only their active parameters per token, which is why some large ones outrun smaller dense models.
Image generators that also fit
These make pictures rather than text. Their memory use scales with the resolution you generate at, not conversation length, so treat these figures as a floor rather than a ceiling.
What you'd need more memory for
The next models up, and what they ask for.
Laguna-M.1needs 76 GB
Qwen3.5-122B-A10Bneeds 77 GB
Mistral-Small-4-119B-2603needs 79 GB
Mistral-Medium-3.5-128Bneeds 80 GB
Devstral-2-123B-Instruct-2512needs 80 GB
Other hardware
Verified against HuggingFace on 2026-09-04New to this? Start here →Runs on a laptop and up · last 12 months only