NVIDIA RTX 4070 (12GB)
The NVIDIA RTX 4070 (12GB) features 12 GB of dedicated VRAM and 504 GB/s memory bandwidth, supporting CUDA and TensorRT-LLM runtimes for full-precision and quantized inference.
Memory
12 GB
Usable Memory Ceiling
12 GB
Bandwidth
504 GB/s
Max Inference Class
14B Models (Q4)
TDP
200W
VRAM Scaling by Context Length
Estimated memory requirements as context length increases. Curves crossing above the reference line indicate Out of Memory (OOM).
Precision:
Best Models to Run on NVIDIA RTX 4070 (12GB)
Inference throughput, memory compatibility, and generation speeds for open-weights LLMs on NVIDIA RTX 4070 (12GB).
Sort by:
Context:
KV Cache:
GPUs:
| Rank | Model | Parameters | Precision | TPS (Approx.) | TTFT | Can Run |
|---|---|---|---|---|---|---|
#112 | Phi-4 Reasoning Plus | 14B | INT4 | 31 tok/s | ~1.6s | |
#125 | Qwen3.5-9B | 9B | INT4 Q8 | 67 tok/s 39 tok/s | ~871 ms ~955 ms | |
#128 | K2 Horizon 7B | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
#135 | Gemma 4 12B | 11.95B | INT4 | 37 tok/s | ~1.3s | |
#140 | Qwen3.5-4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
#144 | GLM-4V | 9B | INT4 | 49 tok/s | ~876 ms | |
#150 | NVIDIA Nemotron 3 Nano 30B-A3B | 3.5B (30B active) | INT4 Q8 FP16 | 22 tok/s 13 tok/s 6 tok/s | ~2.7s ~2.8s ~3.1s | |
#158 | Granite 4.2 8B | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#159 | Gemma 4 E4B | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#166 | Phi-4 | 14B | INT4 | 46 tok/s | ~1.3s | |
#178 | OLMo 3 7B Think | 7B | INT4 Q8 | 59 tok/s 39 tok/s | ~687 ms ~753 ms | |
#179 | DeepSeek-R1 14B | 14B | INT4 | 48 tok/s | ~1.3s | |
#183 | Ministral 3 14B | 14B | INT4 | 31 tok/s | ~1.6s | |
#186 | Gemma 3 12B | 12B | INT4 | 53 tok/s | ~1.2s | |
#189 | Ministral 3 8B | 8B | INT4 Q8 | 53 tok/s 34 tok/s | ~781 ms ~925 ms | |
#191 | MiniCPM5-2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
#192 | Qwen3.5-2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
#194 | Qwen2.5-14B | 14B | INT4 | 46 tok/s | ~1.3s | |
#200 | Gemma 4 E2B | 5.1B | INT4 Q8 FP16 | 103 tok/s 66 tok/s 33 tok/s | ~505 ms ~511 ms ~612 ms | |
#202 | DeepSeek-R1 8B | 8B | INT4 Q8 | 76 tok/s 46 tok/s | ~775 ms ~784 ms | |
#204 | Qwen3-4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
#205 | Ministral 3 3B | 3B | INT4 Q8 FP16 | 98 tok/s 76 tok/s 50 tok/s | ~313 ms ~315 ms ~323 ms | |
#207 | Gemma 2 9B | 9B | INT4 Q8 | 67 tok/s 39 tok/s | ~871 ms ~955 ms | |
#208 | Qwen3.5-0.8B | 0.8B | INT4 Q8 FP16 | 257 tok/s 211 tok/s 151 tok/s | ~92 ms ~93 ms ~95 ms | |
#209 | Qwen3-14B | 14B | INT4 | 46 tok/s | ~1.3s | |
#210 | Gemma 3 4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
#211 | Qwen3-8B | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#213 | DeepSeek-R1 7B | 7B | INT4 Q8 | 84 tok/s 52 tok/s | ~682 ms ~690 ms | |
#214 | Qwen3-1.7B | 1.7B | INT4 Q8 FP16 | 191 tok/s 143 tok/s 91 tok/s | ~186 ms ~188 ms ~192 ms | |
#215 | Llama 3.1 8B | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#216 | Gemma 3 270M | 0.27B | INT4 Q8 FP16 | 295 tok/s 274 tok/s 234 tok/s | ~35 ms ~35 ms ~36 ms | |
#218 | Qwen3-0.6B | 0.6B | INT4 Q8 FP16 | 279 tok/s 238 tok/s 178 tok/s | ~70 ms ~70 ms ~72 ms | |
#219 | Llama 3 8B | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#221 | Phi-4-Mini | 3.8B | INT4 Q8 FP16 | 125 tok/s 83 tok/s 48 tok/s | ~384 ms ~388 ms ~397 ms | |
#223 | Qwen2.5-7B | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
#224 | Ministral-8B-2410 | 8B | INT4 Q8 | 74 tok/s 43 tok/s | ~776 ms ~851 ms | |
#225 | DeepSeek-R1 1.5B | 1.5B | INT4 Q8 FP16 | 203 tok/s 154 tok/s 100 tok/s | ~166 ms ~167 ms ~171 ms | |
#226 | Llama 3.2 3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
#228 | Phi-3-mini | 3.8B | INT4 Q8 FP16 | 125 tok/s 83 tok/s 48 tok/s | ~384 ms ~388 ms ~397 ms | |
#229 | Gemma 3n E2B IT | 6B (2B active) | INT4 Q8 | 97 tok/s 83 tok/s | ~235 ms ~237 ms | |
#230 | Mistral-7B-Instruct-v0.1 | 7.3B | INT4 Q8 | 79 tok/s 49 tok/s | ~710 ms ~718 ms | |
#231 | Llama 3.2 1B | 1B | INT4 Q8 FP16 | 237 tok/s 190 tok/s 132 tok/s | ~115 ms ~116 ms ~119 ms | |
#232 | Gemma 2 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
#233 | Phi-3-medium | 14B | INT4 | 46 tok/s | ~1.3s | |
#234 | OLMo 3 7B Instruct | 7B | INT4 Q8 | 59 tok/s 39 tok/s | ~687 ms ~753 ms | |
#236 | Qwen2-7B | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
#237 | Phi-3-small | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
#238 | Gemma 3 1B | 1B | INT4 Q8 FP16 | 237 tok/s 190 tok/s 132 tok/s | ~115 ms ~116 ms ~119 ms | |
#239 | Mistral-7B-Instruct-v0.2 | 7.3B | INT4 Q8 | 79 tok/s 49 tok/s | ~710 ms ~718 ms | |
#241 | Gemma 1 7B | 7B | INT4 Q8 | 59 tok/s 39 tok/s | ~687 ms ~753 ms | |
#242 | Gemma 1 2B | 2B | INT4 Q8 FP16 | 187 tok/s 134 tok/s 82 tok/s | ~216 ms ~219 ms ~223 ms | |
#243 | ChatGLM3-6B | 6B | INT4 Q8 | 65 tok/s 47 tok/s | ~594 ms ~600 ms | |
#244 | ChatGLM2-6B | 6B | INT4 Q8 | 65 tok/s 47 tok/s | ~594 ms ~600 ms | |
#245 | ChatGLM-6B | 6B | INT4 Q8 | 65 tok/s 47 tok/s | ~594 ms ~600 ms | |
Unranked Models (Sorted by Speed) | ||||||
- | DINOv3 ViT-L/16 | 0.3B | INT4 Q8 FP16 | 319 tok/s 290 tok/s 241 tok/s | ~38 ms ~38 ms ~39 ms | |
- | Qwen2-0.5B | 0.5B | INT4 Q8 FP16 | 292 tok/s 253 tok/s 195 tok/s | ~59 ms ~60 ms ~61 ms | |
- | Qwen2.5-0.5B | 0.5B | INT4 Q8 FP16 | 292 tok/s 253 tok/s 195 tok/s | ~59 ms ~60 ms ~61 ms | |
- | ERNIE-4.5-0.3B | 0.3B | INT4 Q8 FP16 | 287 tok/s 265 tok/s 224 tok/s | ~38 ms ~39 ms ~39 ms | |
- | ERNIE-4.5-0.3B-Base | 0.3B | INT4 Q8 FP16 | 287 tok/s 265 tok/s 224 tok/s | ~38 ms ~39 ms ~39 ms | |
- | Liquid D1-Omni 600M | 0.6B | INT4 Q8 FP16 | 279 tok/s 238 tok/s 178 tok/s | ~70 ms ~70 ms ~72 ms | |
- | PixArt-α XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 278 tok/s 236 tok/s 177 tok/s | ~71 ms ~71 ms ~73 ms | |
- | PixArt-Σ XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 278 tok/s 236 tok/s 177 tok/s | ~71 ms ~71 ms ~73 ms | |
- | Nemotron 3.5 Parse | 0.63B | INT4 Q8 FP16 | 275 tok/s 233 tok/s 174 tok/s | ~73 ms ~74 ms ~75 ms | |
- | SD 2.1 Turbo | 0.86B | INT4 Q8 FP16 | 251 tok/s 204 tok/s 145 tok/s | ~99 ms ~100 ms ~102 ms | |
- | Stable Diffusion 1.4 | 0.86B | INT4 Q8 FP16 | 251 tok/s 204 tok/s 145 tok/s | ~99 ms ~100 ms ~102 ms | |
- | Stable Diffusion 2.1 | 0.86B | INT4 Q8 FP16 | 251 tok/s 204 tok/s 145 tok/s | ~99 ms ~100 ms ~102 ms | |
- | Falcon-1B | 1B | INT4 Q8 FP16 | 248 tok/s 196 tok/s 135 tok/s | ~115 ms ~116 ms ~118 ms | |
- | Falcon3-1B | 1B | INT4 Q8 FP16 | 237 tok/s 190 tok/s 132 tok/s | ~115 ms ~116 ms ~119 ms | |
- | MiniCPM-V 4.7 1B | 1B | INT4 Q8 FP16 | 237 tok/s 190 tok/s 132 tok/s | ~115 ms ~116 ms ~119 ms | |
- | Kandinsky 2.2 Decoder | 1.22B | INT4 Q8 FP16 | 220 tok/s 172 tok/s 115 tok/s | ~137 ms ~139 ms ~142 ms | |
- | Wan 2.1 T2V 1.3B | 1.3B | INT4 Q8 FP16 | 215 tok/s 166 tok/s 111 tok/s | ~145 ms ~146 ms ~149 ms | |
- | Qwen2-1.5B | 1.5B | INT4 Q8 FP16 | 203 tok/s 154 tok/s 100 tok/s | ~166 ms ~167 ms ~171 ms | |
- | Qwen2.5-1.5B | 1.5B | INT4 Q8 FP16 | 203 tok/s 154 tok/s 100 tok/s | ~166 ms ~167 ms ~171 ms | |
- | SVD Image-to-Video XT | 1.5B | INT4 Q8 FP16 | 203 tok/s 154 tok/s 100 tok/s | ~166 ms ~167 ms ~171 ms | |
- | SVD Image-to-Video XT 1.1 | 1.5B | INT4 Q8 FP16 | 203 tok/s 154 tok/s 100 tok/s | ~166 ms ~167 ms ~171 ms | |
- | LFM3 1.7B | 1.7B | INT4 Q8 FP16 | 191 tok/s 143 tok/s 91 tok/s | ~186 ms ~188 ms ~192 ms | |
- | LFM3 1.7B Base | 1.7B | INT4 Q8 FP16 | 191 tok/s 143 tok/s 91 tok/s | ~186 ms ~188 ms ~192 ms | |
- | Spark X2.5 1.7B | 1.7B | INT4 Q8 FP16 | 191 tok/s 143 tok/s 91 tok/s | ~186 ms ~188 ms ~192 ms | |
- | Hy-MT2 1.8B | 1.8B | INT4 Q8 FP16 | 186 tok/s 138 tok/s 87 tok/s | ~196 ms ~198 ms ~202 ms | |
- | CogVideoX 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | LFM3 Audio 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | LFM3 Omni 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | LFM3 VL 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | Liquid D1 2B | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | LTX-Video 0.9.1 | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | Moondream 3.1 | 9B (2B active) | INT4 Q8 | 177 tok/s 123 tok/s | ~243 ms ~266 ms | |
- | Stable Diffusion 3 Medium | 2B | INT4 Q8 FP16 | 177 tok/s 129 tok/s 80 tok/s | ~217 ms ~219 ms ~224 ms | |
- | Kandinsky 3.0 | 2.1B | INT4 Q8 FP16 | 173 tok/s 125 tok/s 78 tok/s | ~226 ms ~228 ms ~233 ms | |
- | LTX-2 | 2.2B | INT4 Q8 FP16 | 169 tok/s 121 tok/s 75 tok/s | ~235 ms ~238 ms ~243 ms | |
- | SDXL 1.0 Refiner | 2.31B | INT4 Q8 FP16 | 165 tok/s 118 tok/s 72 tok/s | ~246 ms ~248 ms ~253 ms | |
- | Stable Diffusion 3.5 Medium | 2.5B | INT4 Q8 FP16 | 158 tok/s 112 tok/s 68 tok/s | ~263 ms ~266 ms ~272 ms | |
- | SDXL 1.0 Base | 2.56B | INT4 Q8 FP16 | 156 tok/s 110 tok/s 66 tok/s | ~269 ms ~271 ms ~277 ms | |
- | SDXL Turbo | 2.56B | INT4 Q8 FP16 | 156 tok/s 110 tok/s 66 tok/s | ~269 ms ~271 ms ~277 ms | |
- | Falcon-3B | 3B | INT4 Q8 FP16 | 151 tok/s 102 tok/s 59 tok/s | ~309 ms ~312 ms ~319 ms | |
- | CroissantLLM Base | 1.3B | INT4 Q8 FP16 | 149 tok/s 126 tok/s 91 tok/s | ~147 ms ~148 ms ~151 ms | |
- | Phi-1 | 1.3B | INT4 Q8 FP16 | 149 tok/s 126 tok/s 91 tok/s | ~147 ms ~148 ms ~151 ms | |
- | Phi-1.5 | 1.3B | INT4 Q8 FP16 | 149 tok/s 126 tok/s 91 tok/s | ~147 ms ~148 ms ~151 ms | |
- | DeepSeek-R1 3B | 3B | INT4 Q8 FP16 | 145 tok/s 99 tok/s 59 tok/s | ~309 ms ~312 ms ~319 ms | |
- | Mellum 2.1 12B A2.5B Thinking | 12B (2.5B active) | INT4 | 145 tok/s | ~302 ms | |
- | Falcon3-3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | LFM2-VL 3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | Liquid D1 3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | Liquid D1-VL 3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | Liquid LFM2-VL 3B D1 | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | Qwen2.5-3B | 3B | INT4 Q8 FP16 | 143 tok/s 99 tok/s 58 tok/s | ~309 ms ~312 ms ~319 ms | |
- | FLUX.2 Klein 4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
- | FrogNano 4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
- | Mage-VL Cowork | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
- | Spark X2.5 4B | 4B | INT4 Q8 FP16 | 121 tok/s 80 tok/s 46 tok/s | ~403 ms ~407 ms ~416 ms | |
- | EmbeddingGemma 2 | 2B | INT4 Q8 FP16 | 116 tok/s 95 tok/s 66 tok/s | ~220 ms ~222 ms ~226 ms | |
- | CogVideoX 1.5 5B | 5B | INT4 Q8 FP16 | 104 tok/s 67 tok/s 36 tok/s | ~496 ms ~501 ms ~556 ms | |
- | CogVideoX 5B | 5B | INT4 Q8 FP16 | 104 tok/s 67 tok/s 36 tok/s | ~496 ms ~501 ms ~556 ms | |
- | Wan 2.2 TI2V 5B | 5B | INT4 Q8 FP16 | 104 tok/s 67 tok/s 36 tok/s | ~496 ms ~501 ms ~556 ms | |
- | Phi-2 | 2.7B | INT4 Q8 FP16 | 102 tok/s 81 tok/s 54 tok/s | ~285 ms ~287 ms ~294 ms | |
- | MaLLaM-3B | 3B | INT4 Q8 FP16 | 98 tok/s 76 tok/s 50 tok/s | ~313 ms ~315 ms ~323 ms | |
- | SmolLM3 3B | 3B | INT4 Q8 FP16 | 98 tok/s 76 tok/s 50 tok/s | ~313 ms ~315 ms ~323 ms | |
- | Falcon-7B | 7B | INT4 Q8 | 85 tok/s 52 tok/s | ~682 ms ~690 ms | |
- | AuraFlow v0.2 | 6.8B | INT4 Q8 | 83 tok/s 52 tok/s | ~664 ms ~671 ms | |
- | AuraFlow v0.3 | 6.8B | INT4 Q8 | 83 tok/s 52 tok/s | ~664 ms ~671 ms | |
- | Falcon3-7B | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
- | Hy-MT2 7B | 7B | INT4 Q8 | 82 tok/s 51 tok/s | ~683 ms ~690 ms | |
- | Mistral-7B-v0.1 | 7.3B | INT4 Q8 | 79 tok/s 49 tok/s | ~710 ms ~718 ms | |
- | Youtu Parsing Omni | 8B | INT4 Q8 | 76 tok/s 46 tok/s | ~775 ms ~784 ms | |
- | Stable Diffusion 3.5 Large | 8.1B | INT4 Q8 | 73 tok/s 43 tok/s | ~785 ms ~861 ms | |
- | Stable Diffusion 3.5 Large Turbo | 8.1B | INT4 Q8 | 73 tok/s 43 tok/s | ~785 ms ~861 ms | |
- | FLUX.2 Klein 9B | 9B | INT4 Q8 | 67 tok/s 39 tok/s | ~871 ms ~955 ms | |
- | LensVLM 9B | 9B | INT4 Q8 | 67 tok/s 39 tok/s | ~871 ms ~955 ms | |
- | MiMo V2.6 Distill Qwen 9B | 9B | INT4 Q8 | 67 tok/s 39 tok/s | ~871 ms ~955 ms | |
- | ChatGLM3-6B-32K | 6B | INT4 Q8 | 65 tok/s 47 tok/s | ~594 ms ~600 ms | |
- | Yi-6B | 6B | INT4 Q8 | 65 tok/s 47 tok/s | ~594 ms ~600 ms | |
- | Falcon3-10B | 10B | INT4 Q8 | 62 tok/s 34 tok/s | ~964 ms ~1.1s | |
- | Mochi 1 Preview | 10B | INT4 Q8 | 62 tok/s 34 tok/s | ~964 ms ~1.1s | |
- | Falcon2-11B | 11B | INT4 Q8 | 59 tok/s 31 tok/s | ~1.1s ~1.3s | |
- | Hunyuan Lite | 7B | INT4 Q8 | 59 tok/s 39 tok/s | ~687 ms ~753 ms | |
- | OLMo 3 7B Base | 7B | INT4 Q8 | 59 tok/s 39 tok/s | ~687 ms ~753 ms | |
- | SEA-LION-7B | 7.1B | INT4 Q8 | 58 tok/s 39 tok/s | ~697 ms ~763 ms | |
- | SEA-LION-7B-Instruct | 7.1B | INT4 Q8 | 58 tok/s 39 tok/s | ~697 ms ~763 ms | |
- | FLUX.1 Canny Dev | 11.9B | INT4 | 53 tok/s | ~1.1s | |
- | FLUX.1 Depth Dev | 11.9B | INT4 | 53 tok/s | ~1.1s | |
- | FLUX.1 Dev | 11.9B | INT4 | 53 tok/s | ~1.1s | |
- | FLUX.1 Fill Dev | 11.9B | INT4 | 53 tok/s | ~1.1s | |
- | FLUX.1 Schnell | 11.9B | INT4 | 53 tok/s | ~1.1s | |
- | Sahabat-AI-Llama3-8B-Instruct | 8B | INT4 Q8 | 53 tok/s 34 tok/s | ~781 ms ~925 ms | |
- | Typhoon-2-8B | 8B | INT4 Q8 | 53 tok/s 34 tok/s | ~781 ms ~925 ms | |
- | GLM-4-9B | 9B | INT4 | 49 tok/s | ~876 ms | |
- | GLM-4-9B-Chat | 9B | INT4 | 49 tok/s | ~876 ms | |
- | GLM-4-9B-Chat-1M | 9B | INT4 | 49 tok/s | ~876 ms | |
- | HunyuanVideo 1.5 720P T2V | 13B | INT4 | 49 tok/s | ~1.2s | |
- | Yi-9B | 9B | INT4 | 49 tok/s | ~876 ms | |
- | Sahabat-AI-Gemma2-9B | 9.2B | INT4 | 48 tok/s | ~895 ms | |
- | Sahabat-AI-Gemma2-9B-Instruct | 9.2B | INT4 | 48 tok/s | ~895 ms | |
- | Wan 2.1 I2V 14B 720P | 14B | INT4 | 46 tok/s | ~1.3s | |
- | Wan 2.1 T2V 14B | 14B | INT4 | 46 tok/s | ~1.3s | |
- | Wan 2.2 T2V A14B | 14B | INT4 | 46 tok/s | ~1.3s | |
- | FLUX.2 Dev | 15B | INT4 | 44 tok/s | ~1.4s | |
- | GreenMind-14B-R1 | 14B | INT4 | 31 tok/s | ~1.6s | |
Inference Capacity
Maximum model parameter class runnable by weight precision.
Context:
4-bit
14B Models
8-bit
8B Models
16-bit
3B Models
Fine-Tuning Capacity
Hardware capacity for local training and adapter fine-tuning.
Context Length:
QLoRA
14B Models
LoRA
8B Models
Full Parameter Training
1B Models
Workload Recommendations
Guidance for optimal precision and operational ceilings on NVIDIA RTX 4070 (12GB).
Inference Sweet Spot
Optimal for 8B models across extended context (32k+). 14B models fit comfortably at Q4 precision with standard context.
Bandwidth & Speed Profile
With 504 GB/s aggregate bandwidth, batch size 1 inference operates in a memory-bandwidth bound regime. Generates approximately 112 tok/s on an 8B Q4 model and 13 tok/s on a 70B Q4 model.
Frequently Asked Questions
©2025 ApX Machine Learning