NVIDIA RTX 4090 (24GB)
The NVIDIA RTX 4090 (24GB) features 24 GB of dedicated VRAM and 1008 GB/s memory bandwidth, supporting CUDA and TensorRT-LLM runtimes for full-precision and quantized inference.
Memory
24 GB
Usable Memory Ceiling
24 GB
Bandwidth
1008 GB/s
Max Inference Class
32B Models (Q4)
TDP
450W
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 4090 (24GB)
Inference throughput, memory compatibility, and generation speeds for open-weights LLMs on NVIDIA RTX 4090 (24GB).
Sort by:
Context:
KV Cache:
GPUs:
| Rank | Model | Parameters | Precision | TPS (Approx.) | TTFT | Can Run |
|---|---|---|---|---|---|---|
#48 | Qwen 3.8 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
#83 | K2 Horizon MoVA 36B A4B | 36B (4B active) | INT4 | 153 tok/s | ~387 ms | |
#93 | Qwen3.5-27B | 27B | INT4 | 48 tok/s | ~1.8s | |
#105 | Sarvam-30B | 32B (2.4B active) | INT4 | 194 tok/s | ~250 ms | |
#106 | Qwen3.6 35B A3B | 35B (3B active) | INT4 | 171 tok/s | ~316 ms | |
#110 | Muse Glimmer 30B | 30B | INT4 | 44 tok/s | ~2.0s | |
#111 | MiMo V2 Flash | 15B (309B active) | INT4 Q8 | 5 tok/s 3 tok/s | ~19.6s ~21.5s | |
#112 | Phi-4 Reasoning Plus | 14B | INT4 Q8 | 62 tok/s 43 tok/s | ~952 ms ~959 ms | |
#117 | Gemma 4 26B A4B | 25.2B (3.8B active) | INT4 | 169 tok/s | ~327 ms | |
#125 | Qwen3.5-9B | 9B | INT4 Q8 FP16 | 113 tok/s 74 tok/s 39 tok/s | ~616 ms ~621 ms ~685 ms | |
#126 | Qwen3.5-35B-A3B | 35B (3B active) | INT4 | 171 tok/s | ~316 ms | |
#128 | K2 Horizon 7B | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
#135 | Gemma 4 12B | 11.95B | INT4 Q8 | 70 tok/s 49 tok/s | ~816 ms ~822 ms | |
#140 | Qwen3.5-4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
#143 | GPT-OSS 20B | 21B (3.6B active) | INT4 | 93 tok/s | ~311 ms | |
#144 | GLM-4V | 9B | INT4 Q8 FP16 | 86 tok/s 62 tok/s 36 tok/s | ~619 ms ~624 ms ~688 ms | |
#150 | NVIDIA Nemotron 3 Nano 30B-A3B | 3.5B (30B active) | INT4 Q8 FP16 | 42 tok/s 25 tok/s 13 tok/s | ~1.9s ~1.9s ~2.0s | |
#153 | Qwen3-30B-A3B | 30B (3B active) | INT4 | 182 tok/s | ~286 ms | |
#158 | Granite 4.2 8B | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#159 | Gemma 4 E4B | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#165 | GLM-4 | 32B | INT4 | 30 tok/s | ~2.5s | |
#166 | Phi-4 | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
#167 | Gemma 3 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
#170 | DeepSeek-R1 32B | 32B | INT4 | 43 tok/s | ~2.1s | |
#171 | Qwen3-32B | 32B | INT4 | 42 tok/s | ~2.1s | |
#175 | Magistral Small | 24B | INT4 | 41 tok/s | ~1.6s | |
#176 | OLMo 3.1 32B Instruct | 32B | INT4 | 30 tok/s | ~2.5s | |
#177 | Qwen2.5-32B | 32B | INT4 | 42 tok/s | ~2.1s | |
#178 | OLMo 3 7B Think | 7B | INT4 Q8 FP16 | 101 tok/s 75 tok/s 46 tok/s | ~486 ms ~490 ms ~498 ms | |
#179 | DeepSeek-R1 14B | 14B | INT4 Q8 | 85 tok/s 52 tok/s | ~948 ms ~955 ms | |
#183 | Ministral 3 14B | 14B | INT4 Q8 | 62 tok/s 43 tok/s | ~952 ms ~959 ms | |
#186 | Gemma 3 12B | 12B | INT4 Q8 FP16 | 92 tok/s 58 tok/s 29 tok/s | ~816 ms ~822 ms ~979 ms | |
#188 | Mistral-Small-2501 | 24B | INT4 | 53 tok/s | ~1.6s | |
#189 | Ministral 3 8B | 8B | INT4 Q8 FP16 | 93 tok/s 68 tok/s 39 tok/s | ~552 ms ~556 ms ~613 ms | |
#190 | Gemma 4 31B | 30.7B | INT4 | 43 tok/s | ~2.0s | |
#191 | MiniCPM5-2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
#192 | Qwen3.5-2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
#193 | Gemma 2 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
#194 | Qwen2.5-14B | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
#195 | OLMo 3.1 32B Think | 32B | INT4 | 30 tok/s | ~2.5s | |
#196 | Nemotron 3.5 Lightning | 30B (3B active) | INT4 | 81 tok/s | ~316 ms | |
#200 | Gemma 4 E2B | 5.1B | INT4 Q8 FP16 | 161 tok/s 112 tok/s 67 tok/s | ~359 ms ~361 ms ~367 ms | |
#202 | DeepSeek-R1 8B | 8B | INT4 Q8 FP16 | 125 tok/s 82 tok/s 46 tok/s | ~549 ms ~553 ms ~563 ms | |
#204 | Qwen3-4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
#205 | Ministral 3 3B | 3B | INT4 Q8 FP16 | 154 tok/s 126 tok/s 88 tok/s | ~222 ms ~223 ms ~227 ms | |
#206 | OLMo 3 32B Base | 32B | INT4 | 30 tok/s | ~2.5s | |
#207 | Gemma 2 9B | 9B | INT4 Q8 FP16 | 113 tok/s 74 tok/s 39 tok/s | ~616 ms ~621 ms ~685 ms | |
#208 | Qwen3.5-0.8B | 0.8B | INT4 Q8 FP16 | 303 tok/s 269 tok/s 215 tok/s | ~67 ms ~68 ms ~69 ms | |
#209 | Qwen3-14B | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
#210 | Gemma 3 4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
#211 | Qwen3-8B | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#213 | DeepSeek-R1 7B | 7B | INT4 Q8 FP16 | 136 tok/s 91 tok/s 52 tok/s | ~484 ms ~487 ms ~495 ms | |
#214 | Qwen3-1.7B | 1.7B | INT4 Q8 FP16 | 252 tok/s 206 tok/s 146 tok/s | ~134 ms ~135 ms ~137 ms | |
#215 | Llama 3.1 8B | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#216 | Gemma 3 270M | 0.27B | INT4 Q8 FP16 | 328 tok/s 315 tok/s 287 tok/s | ~27 ms ~27 ms ~28 ms | |
#217 | Command R | 35B | INT4 | 28 tok/s | ~2.7s | |
#218 | Qwen3-0.6B | 0.6B | INT4 Q8 FP16 | 318 tok/s 289 tok/s 241 tok/s | ~51 ms ~52 ms ~53 ms | |
#219 | Llama 3 8B | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#221 | Phi-4-Mini | 3.8B | INT4 Q8 FP16 | 186 tok/s 136 tok/s 85 tok/s | ~273 ms ~275 ms ~280 ms | |
#223 | Qwen2.5-7B | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
#224 | Ministral-8B-2410 | 8B | INT4 Q8 FP16 | 123 tok/s 81 tok/s 46 tok/s | ~549 ms ~554 ms ~563 ms | |
#225 | DeepSeek-R1 1.5B | 1.5B | INT4 Q8 FP16 | 262 tok/s 217 tok/s 157 tok/s | ~119 ms ~120 ms ~122 ms | |
#226 | Llama 3.2 3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
#227 | Nemotron 3.5 Lightning 30B A3B | 30B (3B active) | INT4 | 182 tok/s | ~286 ms | |
#228 | Phi-3-mini | 3.8B | INT4 Q8 FP16 | 186 tok/s 136 tok/s 85 tok/s | ~273 ms ~275 ms ~280 ms | |
#229 | Gemma 3n E2B IT | 6B (2B active) | INT4 Q8 FP16 | 154 tok/s 136 tok/s 103 tok/s | ~167 ms ~168 ms ~171 ms | |
#230 | Mistral-7B-Instruct-v0.1 | 7.3B | INT4 Q8 FP16 | 130 tok/s 87 tok/s 50 tok/s | ~503 ms ~507 ms ~516 ms | |
#231 | Llama 3.2 1B | 1B | INT4 Q8 FP16 | 289 tok/s 251 tok/s 194 tok/s | ~84 ms ~84 ms ~85 ms | |
#232 | Gemma 2 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
#233 | Phi-3-medium | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
#234 | OLMo 3 7B Instruct | 7B | INT4 Q8 FP16 | 101 tok/s 75 tok/s 46 tok/s | ~486 ms ~490 ms ~498 ms | |
#235 | Yi-34B | 34B | INT4 | 28 tok/s | ~2.6s | |
#236 | Qwen2-7B | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
#237 | Phi-3-small | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
#238 | Gemma 3 1B | 1B | INT4 Q8 FP16 | 289 tok/s 251 tok/s 194 tok/s | ~84 ms ~84 ms ~85 ms | |
#239 | Mistral-7B-Instruct-v0.2 | 7.3B | INT4 Q8 FP16 | 130 tok/s 87 tok/s 50 tok/s | ~503 ms ~507 ms ~516 ms | |
#241 | Gemma 1 7B | 7B | INT4 Q8 FP16 | 101 tok/s 75 tok/s 46 tok/s | ~486 ms ~490 ms ~498 ms | |
#242 | Gemma 1 2B | 2B | INT4 Q8 FP16 | 249 tok/s 197 tok/s 135 tok/s | ~155 ms ~156 ms ~158 ms | |
#243 | ChatGLM3-6B | 6B | INT4 Q8 FP16 | 111 tok/s 83 tok/s 53 tok/s | ~420 ms ~423 ms ~430 ms | |
#244 | ChatGLM2-6B | 6B | INT4 Q8 FP16 | 111 tok/s 83 tok/s 53 tok/s | ~420 ms ~423 ms ~430 ms | |
#245 | ChatGLM-6B | 6B | INT4 Q8 FP16 | 111 tok/s 83 tok/s 53 tok/s | ~420 ms ~423 ms ~430 ms | |
Unranked Models (Sorted by Speed) | ||||||
- | DINOv3 ViT-L/16 | 0.3B | INT4 Q8 FP16 | 343 tok/s 325 tok/s 292 tok/s | ~29 ms ~29 ms ~30 ms | |
- | Qwen2-0.5B | 0.5B | INT4 Q8 FP16 | 326 tok/s 301 tok/s 256 tok/s | ~44 ms ~44 ms ~45 ms | |
- | Qwen2.5-0.5B | 0.5B | INT4 Q8 FP16 | 326 tok/s 301 tok/s 256 tok/s | ~44 ms ~44 ms ~45 ms | |
- | ERNIE-4.5-0.3B | 0.3B | INT4 Q8 FP16 | 323 tok/s 309 tok/s 279 tok/s | ~29 ms ~29 ms ~30 ms | |
- | ERNIE-4.5-0.3B-Base | 0.3B | INT4 Q8 FP16 | 323 tok/s 309 tok/s 279 tok/s | ~29 ms ~29 ms ~30 ms | |
- | Liquid D1-Omni 600M | 0.6B | INT4 Q8 FP16 | 318 tok/s 289 tok/s 241 tok/s | ~51 ms ~52 ms ~53 ms | |
- | PixArt-α XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 318 tok/s 288 tok/s 239 tok/s | ~52 ms ~53 ms ~53 ms | |
- | PixArt-Σ XL 2 1024-MS | 0.61B | INT4 Q8 FP16 | 318 tok/s 288 tok/s 239 tok/s | ~52 ms ~53 ms ~53 ms | |
- | Nemotron 3.5 Parse | 0.63B | INT4 Q8 FP16 | 316 tok/s 286 tok/s 236 tok/s | ~54 ms ~54 ms ~55 ms | |
- | SD 2.1 Turbo | 0.86B | INT4 Q8 FP16 | 299 tok/s 263 tok/s 208 tok/s | ~72 ms ~72 ms ~73 ms | |
- | Stable Diffusion 1.4 | 0.86B | INT4 Q8 FP16 | 299 tok/s 263 tok/s 208 tok/s | ~72 ms ~72 ms ~73 ms | |
- | Stable Diffusion 2.1 | 0.86B | INT4 Q8 FP16 | 299 tok/s 263 tok/s 208 tok/s | ~72 ms ~72 ms ~73 ms | |
- | Falcon-1B | 1B | INT4 Q8 FP16 | 297 tok/s 257 tok/s 198 tok/s | ~83 ms ~84 ms ~85 ms | |
- | Falcon3-1B | 1B | INT4 Q8 FP16 | 289 tok/s 251 tok/s 194 tok/s | ~84 ms ~84 ms ~85 ms | |
- | MiniCPM-V 4.7 1B | 1B | INT4 Q8 FP16 | 289 tok/s 251 tok/s 194 tok/s | ~84 ms ~84 ms ~85 ms | |
- | Kandinsky 2.2 Decoder | 1.22B | INT4 Q8 FP16 | 276 tok/s 235 tok/s 176 tok/s | ~99 ms ~100 ms ~101 ms | |
- | Wan 2.1 T2V 1.3B | 1.3B | INT4 Q8 FP16 | 272 tok/s 230 tok/s 170 tok/s | ~105 ms ~105 ms ~107 ms | |
- | Qwen2-1.5B | 1.5B | INT4 Q8 FP16 | 262 tok/s 217 tok/s 157 tok/s | ~119 ms ~120 ms ~122 ms | |
- | Qwen2.5-1.5B | 1.5B | INT4 Q8 FP16 | 262 tok/s 217 tok/s 157 tok/s | ~119 ms ~120 ms ~122 ms | |
- | SVD Image-to-Video XT | 1.5B | INT4 Q8 FP16 | 262 tok/s 217 tok/s 157 tok/s | ~119 ms ~120 ms ~122 ms | |
- | SVD Image-to-Video XT 1.1 | 1.5B | INT4 Q8 FP16 | 262 tok/s 217 tok/s 157 tok/s | ~119 ms ~120 ms ~122 ms | |
- | LFM3 1.7B | 1.7B | INT4 Q8 FP16 | 252 tok/s 206 tok/s 146 tok/s | ~134 ms ~135 ms ~137 ms | |
- | LFM3 1.7B Base | 1.7B | INT4 Q8 FP16 | 252 tok/s 206 tok/s 146 tok/s | ~134 ms ~135 ms ~137 ms | |
- | Spark X2.5 1.7B | 1.7B | INT4 Q8 FP16 | 252 tok/s 206 tok/s 146 tok/s | ~134 ms ~135 ms ~137 ms | |
- | Hy-MT2 1.8B | 1.8B | INT4 Q8 FP16 | 248 tok/s 201 tok/s 141 tok/s | ~140 ms ~141 ms ~143 ms | |
- | CogVideoX 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | LFM3 Audio 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | LFM3 Omni 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | LFM3 VL 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | Liquid D1 2B | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | LTX-Video 0.9.1 | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | Moondream 3.1 | 9B (2B active) | INT4 Q8 FP16 | 239 tok/s 191 tok/s 126 tok/s | ~174 ms ~175 ms ~192 ms | |
- | Stable Diffusion 3 Medium | 2B | INT4 Q8 FP16 | 239 tok/s 191 tok/s 132 tok/s | ~155 ms ~156 ms ~159 ms | |
- | Kandinsky 3.0 | 2.1B | INT4 Q8 FP16 | 235 tok/s 187 tok/s 128 tok/s | ~162 ms ~163 ms ~165 ms | |
- | LTX-2 | 2.2B | INT4 Q8 FP16 | 232 tok/s 183 tok/s 124 tok/s | ~168 ms ~169 ms ~172 ms | |
- | SDXL 1.0 Refiner | 2.31B | INT4 Q8 FP16 | 228 tok/s 179 tok/s 121 tok/s | ~175 ms ~177 ms ~179 ms | |
- | Stable Diffusion 3.5 Medium | 2.5B | INT4 Q8 FP16 | 222 tok/s 172 tok/s 114 tok/s | ~188 ms ~189 ms ~192 ms | |
- | SDXL 1.0 Base | 2.56B | INT4 Q8 FP16 | 220 tok/s 170 tok/s 113 tok/s | ~192 ms ~193 ms ~196 ms | |
- | SDXL Turbo | 2.56B | INT4 Q8 FP16 | 220 tok/s 170 tok/s 113 tok/s | ~192 ms ~193 ms ~196 ms | |
- | Falcon-3B | 3B | INT4 Q8 FP16 | 215 tok/s 160 tok/s 103 tok/s | ~220 ms ~222 ms ~225 ms | |
- | CroissantLLM Base | 1.3B | INT4 Q8 FP16 | 212 tok/s 188 tok/s 146 tok/s | ~106 ms ~106 ms ~108 ms | |
- | Phi-1 | 1.3B | INT4 Q8 FP16 | 212 tok/s 188 tok/s 146 tok/s | ~106 ms ~106 ms ~108 ms | |
- | Phi-1.5 | 1.3B | INT4 Q8 FP16 | 212 tok/s 188 tok/s 146 tok/s | ~106 ms ~106 ms ~108 ms | |
- | DeepSeek-R1 3B | 3B | INT4 Q8 FP16 | 209 tok/s 157 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | Mellum 2.1 12B A2.5B Thinking | 12B (2.5B active) | INT4 Q8 FP16 | 209 tok/s 164 tok/s 100 tok/s | ~215 ms ~217 ms ~257 ms | |
- | Falcon3-3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | LFM2-VL 3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | Liquid D1 3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | Liquid D1-VL 3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | Liquid LFM2-VL 3B D1 | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | Qwen2.5-3B | 3B | INT4 Q8 FP16 | 207 tok/s 156 tok/s 101 tok/s | ~220 ms ~222 ms ~225 ms | |
- | LFM2.5 24B A2B | 24B (2.5B active) | INT4 | 197 tok/s | ~243 ms | |
- | ERNIE-4.5-21B-A3B | 21B (3B active) | INT4 Q8 | 188 tok/s 131 tok/s | ~268 ms ~315 ms | |
- | ERNIE-4.5-21B-A3B-Base | 21B (3B active) | INT4 Q8 | 188 tok/s 131 tok/s | ~268 ms ~315 ms | |
- | ERNIE-4.5-VL-28B-A3B | 28B (3B active) | INT4 | 183 tok/s | ~281 ms | |
- | ERNIE-4.5-VL-28B-A3B-Base | 28B (3B active) | INT4 | 183 tok/s | ~281 ms | |
- | FLUX.2 Klein 4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
- | FrogNano 4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
- | Hy-Memory 30B A3B | 30B (3B active) | INT4 | 182 tok/s | ~286 ms | |
- | Hy-MT2 30B A3B | 30B (3B active) | INT4 | 182 tok/s | ~286 ms | |
- | Mage-VL Cowork | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
- | Spark X2.5 4B | 4B | INT4 Q8 FP16 | 182 tok/s 131 tok/s 81 tok/s | ~286 ms ~288 ms ~293 ms | |
- | EmbeddingGemma 2 | 2B | INT4 Q8 FP16 | 176 tok/s 151 tok/s 112 tok/s | ~157 ms ~158 ms ~160 ms | |
- | MiniCPM-V 4.7 35B A3B | 35B (3B active) | INT4 | 171 tok/s | ~316 ms | |
- | CogVideoX 1.5 5B | 5B | INT4 Q8 FP16 | 162 tok/s 114 tok/s 68 tok/s | ~352 ms ~355 ms ~361 ms | |
- | CogVideoX 5B | 5B | INT4 Q8 FP16 | 162 tok/s 114 tok/s 68 tok/s | ~352 ms ~355 ms ~361 ms | |
- | Wan 2.2 TI2V 5B | 5B | INT4 Q8 FP16 | 162 tok/s 114 tok/s 68 tok/s | ~352 ms ~355 ms ~361 ms | |
- | Phi-2 | 2.7B | INT4 Q8 FP16 | 160 tok/s 133 tok/s 94 tok/s | ~202 ms ~204 ms ~207 ms | |
- | MaLLaM-3B | 3B | INT4 Q8 FP16 | 154 tok/s 126 tok/s 88 tok/s | ~222 ms ~223 ms ~227 ms | |
- | SmolLM3 3B | 3B | INT4 Q8 FP16 | 154 tok/s 126 tok/s 88 tok/s | ~222 ms ~223 ms ~227 ms | |
- | Falcon-7B | 7B | INT4 Q8 FP16 | 138 tok/s 91 tok/s 52 tok/s | ~483 ms ~487 ms ~495 ms | |
- | AuraFlow v0.2 | 6.8B | INT4 Q8 FP16 | 136 tok/s 91 tok/s 53 tok/s | ~471 ms ~474 ms ~482 ms | |
- | AuraFlow v0.3 | 6.8B | INT4 Q8 FP16 | 136 tok/s 91 tok/s 53 tok/s | ~471 ms ~474 ms ~482 ms | |
- | Falcon3-7B | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
- | Hy-MT2 7B | 7B | INT4 Q8 FP16 | 134 tok/s 89 tok/s 52 tok/s | ~484 ms ~487 ms ~496 ms | |
- | Mistral-7B-v0.1 | 7.3B | INT4 Q8 FP16 | 130 tok/s 87 tok/s 50 tok/s | ~503 ms ~507 ms ~516 ms | |
- | Youtu Parsing Omni | 8B | INT4 Q8 FP16 | 125 tok/s 82 tok/s 46 tok/s | ~549 ms ~553 ms ~563 ms | |
- | Stable Diffusion 3.5 Large | 8.1B | INT4 Q8 FP16 | 122 tok/s 80 tok/s 46 tok/s | ~556 ms ~560 ms ~570 ms | |
- | Stable Diffusion 3.5 Large Turbo | 8.1B | INT4 Q8 FP16 | 122 tok/s 80 tok/s 46 tok/s | ~556 ms ~560 ms ~570 ms | |
- | FLUX.2 Klein 9B | 9B | INT4 Q8 FP16 | 113 tok/s 74 tok/s 39 tok/s | ~616 ms ~621 ms ~685 ms | |
- | LensVLM 9B | 9B | INT4 Q8 FP16 | 113 tok/s 74 tok/s 39 tok/s | ~616 ms ~621 ms ~685 ms | |
- | MiMo V2.6 Distill Qwen 9B | 9B | INT4 Q8 FP16 | 113 tok/s 74 tok/s 39 tok/s | ~616 ms ~621 ms ~685 ms | |
- | ChatGLM3-6B-32K | 6B | INT4 Q8 FP16 | 111 tok/s 83 tok/s 53 tok/s | ~420 ms ~423 ms ~430 ms | |
- | Yi-6B | 6B | INT4 Q8 FP16 | 111 tok/s 83 tok/s 53 tok/s | ~420 ms ~423 ms ~430 ms | |
- | Falcon3-10B | 10B | INT4 Q8 FP16 | 106 tok/s 68 tok/s 36 tok/s | ~682 ms ~688 ms ~758 ms | |
- | Kimi-VL-A3B-Instruct | 16B (3B active) | INT4 Q8 | 106 tok/s 89 tok/s | ~262 ms ~285 ms | |
- | Mochi 1 Preview | 10B | INT4 Q8 FP16 | 106 tok/s 68 tok/s 36 tok/s | ~682 ms ~688 ms ~758 ms | |
- | Falcon2-11B | 11B | INT4 Q8 FP16 | 102 tok/s 64 tok/s 32 tok/s | ~748 ms ~754 ms ~898 ms | |
- | Hunyuan Lite | 7B | INT4 Q8 FP16 | 101 tok/s 75 tok/s 46 tok/s | ~486 ms ~490 ms ~498 ms | |
- | OLMo 3 7B Base | 7B | INT4 Q8 FP16 | 101 tok/s 75 tok/s 46 tok/s | ~486 ms ~490 ms ~498 ms | |
- | SEA-LION-7B | 7.1B | INT4 Q8 FP16 | 101 tok/s 74 tok/s 46 tok/s | ~493 ms ~496 ms ~504 ms | |
- | SEA-LION-7B-Instruct | 7.1B | INT4 Q8 FP16 | 101 tok/s 74 tok/s 46 tok/s | ~493 ms ~496 ms ~504 ms | |
- | FLUX.1 Canny Dev | 11.9B | INT4 Q8 FP16 | 93 tok/s 59 tok/s 29 tok/s | ~809 ms ~815 ms ~972 ms | |
- | FLUX.1 Depth Dev | 11.9B | INT4 Q8 FP16 | 93 tok/s 59 tok/s 29 tok/s | ~809 ms ~815 ms ~972 ms | |
- | FLUX.1 Dev | 11.9B | INT4 Q8 FP16 | 93 tok/s 59 tok/s 29 tok/s | ~809 ms ~815 ms ~972 ms | |
- | FLUX.1 Fill Dev | 11.9B | INT4 Q8 FP16 | 93 tok/s 59 tok/s 29 tok/s | ~809 ms ~815 ms ~972 ms | |
- | FLUX.1 Schnell | 11.9B | INT4 Q8 FP16 | 93 tok/s 59 tok/s 29 tok/s | ~809 ms ~815 ms ~972 ms | |
- | Sahabat-AI-Llama3-8B-Instruct | 8B | INT4 Q8 FP16 | 93 tok/s 68 tok/s 39 tok/s | ~552 ms ~556 ms ~613 ms | |
- | Typhoon-2-8B | 8B | INT4 Q8 FP16 | 93 tok/s 68 tok/s 39 tok/s | ~552 ms ~556 ms ~613 ms | |
- | HunyuanVideo 1.5 720P T2V | 13B | INT4 Q8 | 87 tok/s 54 tok/s | ~882 ms ~889 ms | |
- | GLM-4-9B | 9B | INT4 Q8 FP16 | 86 tok/s 62 tok/s 36 tok/s | ~619 ms ~624 ms ~688 ms | |
- | GLM-4-9B-Chat | 9B | INT4 Q8 FP16 | 86 tok/s 62 tok/s 36 tok/s | ~619 ms ~624 ms ~688 ms | |
- | GLM-4-9B-Chat-1M | 9B | INT4 Q8 FP16 | 86 tok/s 62 tok/s 36 tok/s | ~619 ms ~624 ms ~688 ms | |
- | Yi-9B | 9B | INT4 Q8 FP16 | 86 tok/s 62 tok/s 36 tok/s | ~619 ms ~624 ms ~688 ms | |
- | Sahabat-AI-Gemma2-9B | 9.2B | INT4 Q8 FP16 | 85 tok/s 61 tok/s 35 tok/s | ~632 ms ~637 ms ~702 ms | |
- | Sahabat-AI-Gemma2-9B-Instruct | 9.2B | INT4 Q8 FP16 | 85 tok/s 61 tok/s 35 tok/s | ~632 ms ~637 ms ~702 ms | |
- | Wan 2.1 I2V 14B 720P | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
- | Wan 2.1 T2V 14B | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
- | Wan 2.2 T2V A14B | 14B | INT4 Q8 | 82 tok/s 51 tok/s | ~948 ms ~955 ms | |
- | FLUX.2 Dev | 15B | INT4 Q8 | 78 tok/s 48 tok/s | ~1.0s ~1.0s | |
- | GreenMind-14B-R1 | 14B | INT4 Q8 | 62 tok/s 43 tok/s | ~952 ms ~959 ms | |
- | Arete 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
- | AREX-2 | 27B | INT4 | 48 tok/s | ~1.8s | |
- | PPLX Decider v1 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
- | PPLX Decider v1.1 27B | 27B | INT4 | 48 tok/s | ~1.8s | |
- | Agnes 3.0 Flash | 33B | INT4 | 41 tok/s | ~2.2s | |
- | Falcon-40B | 40B | INT4 | 34 tok/s | ~2.9s | |
Inference Capacity
Maximum model parameter class runnable by weight precision.
Context:
4-bit
32B Models
8-bit
14B Models
16-bit
8B Models
Fine-Tuning Capacity
Hardware capacity for local training and adapter fine-tuning.
Context Length:
QLoRA
32B Models
LoRA
14B Models
Full Parameter Training
3B Models
Workload Recommendations
Guidance for optimal precision and operational ceilings on NVIDIA RTX 4090 (24GB).
Inference Sweet Spot
Optimized for 14B models at uncompressed or Q8 precision, and 32B models at Q4 with up to 32k context.
Bandwidth & Speed Profile
With 1008 GB/s aggregate bandwidth, batch size 1 inference operates in a memory-bandwidth bound regime. Generates approximately 224 tok/s on an 8B Q4 model and 27 tok/s on a 70B Q4 model.
Frequently Asked Questions
©2025 ApX Machine Learning