Atom 1 128GB

Linux Ultra tier · 14-core ARM v9 + NVIDIA GPU · 128 GB RAM · 273 GB/s bandwidth
View device page →
BENCHMARK RESULTS (74)
Model Quant Measured Estimated RAM used App Source Date
Ternary Bonsai 4B Ternary Bonsai 4B Q2 220 tok/s 200 tok/s +10% 1.1 GB Ollama community 2026-01-17
🗣️ OuteTTS 0.3 500M FP16 220 tok/s 200 tok/s +10% 1.2 GB Piper community 2026-01-14
🔢 GTE Large FP16 219 tok/s 200 tok/s +10% 0.8 GB Xybrid CLI community 2026-03-16
🗣️ Kokoro 82M FP16 219 tok/s 200 tok/s +10% 0.4 GB Piper community 2026-02-22
SmolLM2 360M SmolLM2 360M FP16 217 tok/s 200 tok/s +9% 1.0 GB Xybrid CLI community 2026-01-03
🔢 BGE Small FP16 215 tok/s 200 tok/s +8% 0.1 GB Ollama community 2026-03-02
Qwen 2.5 0.5B Qwen 2.5 0.5B FP16 210 tok/s 200 tok/s +5% 1.5 GB LM Studio editorial 2026-02-24
Whisper Tiny Whisper Tiny FP16 208 tok/s 200 tok/s +4% 0.1 GB Whisper Transcription community 2026-01-04
Ternary Bonsai 1.7B Ternary Bonsai 1.7B Q2 205 tok/s 200 tok/s +3% 0.5 GB Jan community 2026-01-02
👁️ SmolVLM 500M FP16 204 tok/s 200 tok/s +2% 1.4 GB Jan community 2026-04-20
Qwen 2.5 Coder 0.5B Qwen 2.5 Coder 0.5B FP16 203 tok/s 200 tok/s +2% 1.3 GB Ollama community 2026-02-10
Bonsai 8B (1-bit) Bonsai 8B (1-bit) Q1 202 tok/s 200 tok/s +1% 1.5 GB Jan community 2026-05-25
Whisper Small Whisper Small FP16 202 tok/s 200 tok/s +1% 0.6 GB Xybrid CLI community 2026-03-28
🗣️ KittenTTS Mini FP16 198 tok/s 200 tok/s -1% 0.2 GB Piper community 2026-01-01
🔢 Nomic Embed Text FP16 190 tok/s 200 tok/s -5% 0.3 GB Ollama editorial 2026-01-05
🔢 all-MiniLM-L6-v2 FP16 186 tok/s 200 tok/s -7% 0.1 GB Xybrid CLI community 2026-05-13
🎙️ Distil-Whisper Large V3 FP16 183 tok/s 182 tok/s +1% 1.7 GB Whisper Transcription community 2026-02-06
SmolLM2 135M SmolLM2 135M FP16 181 tok/s 200 tok/s -9% 0.3 GB Jan community 2026-05-22
🗣️ KittenTTS Nano FP16 178 tok/s 200 tok/s -11% 0.1 GB Xybrid CLI community 2026-05-14
Wav2Vec2 Base Wav2Vec2 Base FP16 177 tok/s 200 tok/s -11% 0.3 GB Xybrid CLI community 2026-03-05
Ternary Bonsai 8B Ternary Bonsai 8B Q2 171 tok/s 171 tok/s +0% 1.9 GB Ollama community 2026-02-20
🗣️ NeuTTS Air FP16 167 tok/s 182 tok/s -8% 1.8 GB Xybrid CLI community 2026-04-09
Whisper Medium Whisper Medium FP16 167 tok/s 182 tok/s -8% 1.9 GB Xybrid CLI editorial 2026-01-01
Bonsai Image 4B Bonsai Image 4B Q2 151 tok/s 152 tok/s -1% 2.2 GB Ollama community 2026-05-24
Qwen 3.5 0.8B Qwen 3.5 0.8B FP16 139 tok/s 152 tok/s -9% 2.1 GB Jan community 2026-05-14
🎨 Stable Diffusion Turbo FP16 135 tok/s 130 tok/s +4% 2.2 GB LM Studio community 2026-02-25
TinyLlama 1.1B TinyLlama 1.1B FP16 128 tok/s 124 tok/s +3% 2.5 GB Jan community 2026-03-15
Gemma 3 1B Gemma 3 1B FP16 122 tok/s 124 tok/s -2% 2.7 GB Xybrid CLI editorial 2026-01-22
LFM2.5 1.2B LFM2.5 1.2B FP16 113 tok/s 117 tok/s -3% 3.0 GB LM Studio community 2026-01-04
Llama 3.2 1B Llama 3.2 1B FP16 94 tok/s 101 tok/s -7% 3.4 GB LM Studio editorial 2026-05-27
🧠 DeepSeek R1 Distill 1.5B FP16 88 tok/s 83 tok/s +6% 3.6 GB Xybrid CLI community 2026-01-06
Whisper Large V3 Whisper Large V3 FP16 88 tok/s 88 tok/s +0% 3.4 GB Xybrid CLI community 2026-01-02
StableLM 2 1.6B StableLM 2 1.6B FP16 83 tok/s 83 tok/s +0% 4.0 GB LM Studio community 2026-02-27
🗣️ Dia 1.6B FP16 83 tok/s 83 tok/s +0% 4.1 GB Piper community 2026-01-22
SmolLM2 1.7B SmolLM2 1.7B FP16 81 tok/s 80 tok/s +1% 3.9 GB LM Studio editorial 2026-04-07
👁️ Moondream 2B FP16 77 tok/s 74 tok/s +4% 4.2 GB LM Studio editorial 2026-01-16
Qwen 2.5 Coder 1.5B Qwen 2.5 Coder 1.5B FP16 75 tok/s 85 tok/s -12% 3.9 GB Jan community 2026-02-20
Qwen 3.5 2B Qwen 3.5 2B FP16 63 tok/s 65 tok/s -3% 4.4 GB Ollama community 2026-05-27
🎨 Stable Diffusion 3.5 Medium FP16 60 tok/s 59 tok/s +2% 5.8 GB Ollama community 2026-01-21
Gemma 4 E2B Gemma 4 E2B FP16 59 tok/s 61 tok/s -3% 5.6 GB Ollama community 2026-03-15
Qwen 2.5 Coder 3B Qwen 2.5 Coder 3B FP16 44 tok/s 43 tok/s +2% 7.1 GB Jan community 2026-03-14
Llama 3.2 3B Llama 3.2 3B FP16 43 tok/s 41 tok/s +5% 7.8 GB Jan community 2026-04-27
👨‍💻 StarCoder2 3B FP16 41 tok/s 44 tok/s -7% 7.4 GB Ollama community 2026-04-19
🎨 SDXL Turbo FP16 41 tok/s 42 tok/s -2% 7.8 GB Xybrid CLI community 2026-02-25
Qwen 2.5 3B Qwen 2.5 3B FP16 39 tok/s 43 tok/s -9% 7.2 GB Jan community 2026-04-10
Phi-4 Mini Phi-4 Mini FP16 35 tok/s 35 tok/s +0% 9.6 GB Jan community 2026-04-19
Gemma 4 E4B Gemma 4 E4B FP16 32 tok/s 30 tok/s +7% 9.9 GB Xybrid CLI community 2026-02-08
Qwen 3.5 4B Qwen 3.5 4B FP16 31 tok/s 33 tok/s -6% 8.6 GB LM Studio community 2026-01-03
Gemma 3n E2B Gemma 3n E2B FP16 30 tok/s 31 tok/s -3% 11.5 GB LM Studio editorial 2026-01-02
Gemma 3 4B Gemma 3 4B FP16 29 tok/s 31 tok/s -6% 10.3 GB LM Studio editorial 2026-03-16
👨‍💻 DeepSeek Coder 6.7B FP16 21 tok/s 20 tok/s +5% 17.8 GB Ollama editorial 2026-04-07
Gemma 3n E4B Gemma 3n E4B FP16 20 tok/s 20 tok/s +0% 15.1 GB Jan community 2026-01-11
Mistral 7B Mistral 7B FP16 19 tok/s 19 tok/s +0% 16.7 GB Jan community 2026-02-07
Qwen 2.5 Coder 7B Qwen 2.5 Coder 7B FP16 19 tok/s 18 tok/s +6% 17.6 GB Jan community 2026-02-11
👁️ LLaVA 1.6 7B FP16 19 tok/s 19 tok/s +0% 16.4 GB Ollama community 2026-01-05
Qwen 2.5 7B Qwen 2.5 7B FP16 17 tok/s 18 tok/s -6% 18.2 GB Xybrid CLI community 2026-01-21
Llama 3.1 8B Llama 3.1 8B FP16 17 tok/s 17 tok/s +0% 17.8 GB Ollama community 2026-02-09
🧠 DeepSeek R1 Distill 7B FP16 17 tok/s 18 tok/s -6% 16.8 GB Ollama community 2026-01-11
🧠 DeepSeek R1 Distill 8B FP16 16 tok/s 17 tok/s -6% 19.9 GB Ollama community 2026-04-21
Qwen 2.5 VL 7B Qwen 2.5 VL 7B FP16 16 tok/s 17 tok/s -6% 17.6 GB Xybrid CLI community 2026-04-16
LFM2.5 8B A1B LFM2.5 8B A1B FP16 15 tok/s 16 tok/s -6% 18.8 GB Jan community 2026-05-13
Qwen 3.5 9B Qwen 3.5 9B FP16 14 tok/s 15 tok/s -7% 23.5 GB Jan community 2026-03-27
🧠 Qwen3 8B FP16 14 tok/s 16 tok/s -12% 18.7 GB LM Studio editorial 2026-01-06
Gemma 3 12B Gemma 3 12B FP16 12 tok/s 11 tok/s +9% 28.5 GB Ollama community 2026-04-06
Mistral Nemo 12B Mistral Nemo 12B FP16 11 tok/s 11 tok/s +0% 31.6 GB LM Studio editorial 2026-05-25
Phi-4 Medium Phi-4 Medium FP16 10 tok/s 10 tok/s +0% 35.3 GB Ollama community 2026-03-17
🎨 FLUX.1 Schnell FP16 10 tok/s 11 tok/s -9% 28.6 GB Xybrid CLI community 2026-01-19
Gemma 4 26B A4B Gemma 4 26B A4B FP16 5 tok/s 5 tok/s +0% 63.4 GB Xybrid CLI community 2026-02-23
Gemma 4 31B Gemma 4 31B FP16 4 tok/s 4 tok/s +0% 77.1 GB Xybrid CLI community 2026-01-21
Qwen 3.5 35B A3B Qwen 3.5 35B A3B FP16 4 tok/s 4 tok/s +0% 82.8 GB Ollama community 2026-02-12
👨‍💻 Laguna XS.2 FP16 4 tok/s 4 tok/s +0% 81.0 GB LM Studio community 2026-02-26
Llama 4 Scout Llama 4 Scout Q6 3 tok/s 3 tok/s +0% 93.5 GB Ollama community 2026-03-01
DeepSeek V4 Flash DeepSeek V4 Flash Q2 3 tok/s 3 tok/s +0% 110.4 GB Jan community 2026-03-15
Qwen3 Coder Next Qwen3 Coder Next Q8 3 tok/s 3 tok/s +0% 104.8 GB Xybrid CLI community 2026-05-23
← All benchmarks