Best local AI models for 128GB VRAM
Ranked by what actually fits at 32K context, computed from real file bytes.
A 128GB card gives you about 119.04 GiB to work with after driver overhead. 2091 indexed models fit at 32K context — the largest being MiniMax-M3 at 427B parameters in IQ2_XXS.
From the file· fit from summed bytesFrom the file· KV per layer
alltext generation 1796vision + language 191speech recognition 39speech synthesis 21embeddings 26video generation 16image generation 2
Fits in 128GB at 32K context
largest quantization that fits, per model
| Model | Modality | Best quant | Params○ | Total◐ | Headroom◐ |
|---|---|---|---|---|---|
| Qwen3-Coder-30B-A3B-InstructMoE | text generation | BF16 | 30.5B | 60.69 GiB | 58.35 GiB |
| Qwen3.6-27B | text generation | BF16 | 27.8B | 53.77 GiB | 65.27 GiB |
| Qwen3.6-35B-A3BMoE | vision + language | BF16 | 36.0B | 67.61 GiB | 51.43 GiB |
| Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP | vision + language | Q6_K | 27.8B | 90.87 GiB | 28.17 GiB |
| Qwen3.8-27B | text generation | BF16 | 27.8B | 58.51 GiB | 60.53 GiB |
| Qwen3.5-9B | vision + language | BF16 | 9.7B | 18.98 GiB | 100.06 GiB |
| gemma-4-26B-A4B-itMoE | vision + language | BF16 | 26.5B | 50.98 GiB | 68.06 GiB |
| gemma-4-12B-it | vision + language | BF16 | 12.0B | 25.51 GiB | 93.53 GiB |
| DeepSeek-V4-FlashMoE | text generation | UD-IQ3_S | 291B | 110.16 GiB | 8.88 GiB |
| nemotron-3.5-asr-streaming-0.6b | speech recognition | F32 | 638M | 3.22 GiB | 115.82 GiB |
| Qwen3.5-4B | vision + language | BF16 | 4.7B | 9.88 GiB | 109.16 GiB |
| gemma-4-12B-it-qat-q4_0-unquantized | text generation | Q4_0 | 12.0B | 9.81 GiB | 109.23 GiB |
| Qwythos-9B-Claude-Mythos-5-1M | vision + language | BF16 | 9.4B | 35.67 GiB | 83.37 GiB |
| gemma-4-E4B-it | text generation | BF16 | 8.0B | 15.50 GiB | 103.54 GiB |
| Qwen3-30B-A3B-Thinking-2507MoE | text generation | BF16 | 30.5B | 60.69 GiB | 58.35 GiB |
| gemma-4-31B-it | vision + language | BF16 | 31.3B | 64.25 GiB | 54.79 GiB |
| Muse-Glimmer-30B | vision + language | BF16 | 29.8B | 53.28 GiB | 65.76 GiB |
| Qwen3-4B | text generation | BF16 | 4.0B | 12.81 GiB | 106.23 GiB |
| Qwen3-8B | text generation | BF16 | 8.2B | 20.60 GiB | 98.44 GiB |
| gemma-4-26B-A4B-it-qat-q4_0-unquantizedMoE | vision + language | Q4_0 | 26.5B | 15.78 GiB | 103.26 GiB |
| Qwen3.5-122B-A10BMoE | vision + language | Q6_K_L | 125B | 102.88 GiB | 16.16 GiB |
| Laguna-XS-2.1MoE | text generation | BF16 | 33.4B | 64.50 GiB | 54.54 GiB |
| DeepSeek-V4-Flash-0731MoE | text generation | UD-IQ3_S | 304B | 109.01 GiB | 10.03 GiB |
| Qwen3.5-0.8B | vision + language | BF16 | 873M | 2.60 GiB | 116.44 GiB |
| Llama-3.2-1B-Instruct | text generation | F16 | 1.2B | 4.11 GiB | 114.93 GiB |
| Qwen-AgentWorld-35B-A3BMoE | text generation | BF16 | 34.7B | 67.62 GiB | 51.42 GiB |
| gemma-4-E2B-it | vision + language | BF16 | 5.1B | 9.71 GiB | 109.33 GiB |
| gemma-4-31B-it-qat-q4_0-unquantized | vision + language | Q4_0 | 32.7B | 23.49 GiB | 95.55 GiB |
| gpt-oss-20bMoE | text generation | F16 | 21.5B | 14.40 GiB | 104.64 GiB |
| KAT-Coder-V2.5-DevMoE | text generation | BF16 | 34.7B | 66.04 GiB | 53.00 GiB |
| gemma-4-E4B-it-qat-q4_0-unquantized | vision + language | Q4_0 | 7.9B | 6.12 GiB | 112.92 GiB |
| Qwen3-VL-30B-A3B-InstructMoE | vision + language | BF16 | 31.1B | 60.69 GiB | 58.35 GiB |
| Qwen3-30B-A3BMoE | text generation | BF16 | 30.5B | 60.69 GiB | 58.35 GiB |
| Qwopus3.6-35B-A3B-v1MoE | vision + language | F16 | 36.0B | 67.61 GiB | 51.43 GiB |
| llama-3-youko-8b | text generation | Q8_0 | 8.0B | 12.79 GiB | 106.25 GiB |
| Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP | vision + language | Q8_0 | 9.7B | 21.60 GiB | 97.44 GiB |
| Laguna-S-2.1MoE | text generation | Q6_K_L | 118B | 97.30 GiB | 21.74 GiB |
| parakeet-tdt-0.6b-v3 | speech recognition | F32 | 627M | 3.18 GiB | 115.86 GiB |
| Qwen3.5-35B-A3BMoE | vision + language | BF16 | 36.0B | 67.61 GiB | 51.43 GiB |
| Llama-3.1-8B-Instruct | text generation | F32 | 8.0B | 34.76 GiB | 84.28 GiB |
| ced-base | text generation | F32 | 86M | 1.16 GiB | 117.88 GiB |
| Hy3MoE | text generation | Q2_K | 299B | 112.12 GiB | 6.92 GiB |
| Ace-Step1.5 | speech synthesis | BF16 | 160M | 84.36 GiB | 34.68 GiB |
| Qwen2.5-7B-Instruct | text generation | F16 | 7.6B | 16.80 GiB | 102.24 GiB |
| Qwen3-TTS-12Hz-0.6B-Base | speech synthesis | F32 | 915M | 29.72 GiB | 89.32 GiB |
| Qwythos-9B-v2 | vision + language | BF16 | 9.7B | 35.67 GiB | 83.37 GiB |
| UI-TARS-1.5-7B | text generation | F16 | 8.3B | 16.80 GiB | 102.24 GiB |
| gemma-3-1b-it | text generation | F16 | 1000M | 2.81 GiB | 116.23 GiB |
| gemma-4-E2B-it-qat-q4_0-unquantized | text generation | BF16 | 5.1B | 9.83 GiB | 109.21 GiB |
| Qwen3-1.7B | text generation | BF16 | 2.0B | 8.08 GiB | 110.96 GiB |
| GLM-4.7-FlashMoE | text generation | BF16 | 31.2B | 58.26 GiB | 60.78 GiB |
| whisper-medium | speech recognition | F32 | 764M | 3.69 GiB | 115.35 GiB |
| embeddinggemma-300m | text generation | F32 | 303M | 2.05 GiB | 116.99 GiB |
| Llama-3.2-3B-Instruct | text generation | F16 | 3.2B | 10.30 GiB | 108.74 GiB |
| Qwen3-0.6B | text generation | BF16 | 752M | 5.68 GiB | 113.36 GiB |
| Qwen3-14B | text generation | BF16 | 14.8B | 33.37 GiB | 85.67 GiB |
| Ornith-1.0-35BMoE | text generation | BF16 | 34.7B | 67.62 GiB | 51.42 GiB |
| ThinkingCap-Qwen3.6-27B | vision + language | F16 | 27.4B | 53.77 GiB | 65.27 GiB |
| Qwopus3.6-27B-Coder | vision + language | Q8_0 | 27.8B | 29.92 GiB | 89.12 GiB |
| Voxtral-Mini-4B-Realtime-2602 | speech recognition | F16 | 4.4B | 12.33 GiB | 106.71 GiB |
This page models a generic 128GB accelerator, so it answers what fits rather than how fast it runs. For tokens per second you need a specific card — pick one from hardware, where bandwidth is known.