NVIDIA A100 (40GB)
The NVIDIA A100 (40GB) features 40 GB of dedicated VRAM and 1555 GB/s memory bandwidth, supporting CUDA and TensorRT-LLM runtimes for full-precision and quantized inference.
Memory
40 GB
Usable Memory Ceiling
40 GB
Bandwidth
1555 GB/s
Max Inference Class
32B Models (Q4)
TDP
400W
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 A100 (40GB)
Inference throughput, memory compatibility, and generation speeds for open-weights LLMs on NVIDIA A100 (40GB).
Sort by:
Context:
KV Cache:
GPUs:
| Rank | Model | Parameters | Precision | TPS (Approx.) | TTFT | Can Run |
|---|---|---|---|---|---|---|
#33 | Qwen 3.8 27B | 27B | INT4 Q8 | 72 tok/s 42 tok/s | ~1.1s ~1.1s | |
#48 | Agnes 3.0 Flash | 33B | INT4 Q8 | 60 tok/s 33 tok/s | ~1.3s ~1.4s | |
#72 | K2 Horizon MoVA 36B A4B | 36B (4B active) | INT4 Q8 | 255 tok/s 178 tok/s | ~214 ms ~234 ms | |
#83 | Qwen3.5-27B | 27B | INT4 Q8 | 72 tok/s 42 tok/s | ~1.1s ~1.1s | |
#92 | Sarvam-30B | 32B (2.4B active) | INT4 Q8 | 314 tok/s 237 tok/s | ~150 ms ~163 ms | |
#94 | Qwen3.6 35B A3B | 35B (3B active) | INT4 Q8 | 289 tok/s 210 tok/s | ~175 ms ~190 ms | |
#96 | Muse Glimmer 30B | 30B | INT4 Q8 | 66 tok/s 36 tok/s | ~1.2s ~1.3s | |
#100 | Phi-4 Reasoning Plus | 14B | INT4 Q8 FP16 | 93 tok/s 64 tok/s 35 tok/s | ~572 ms ~577 ms ~639 ms | |
#101 | MiMo V2 Flash | 15B (309B active) | INT4 Q8 FP16 | 7 tok/s 4 tok/s 2 tok/s | ~11.8s ~11.9s ~13.2s | |
#108 | Gemma 4 26B A4B | 25.2B (3.8B active) | INT4 Q8 | 270 tok/s 198 tok/s | ~196 ms ~197 ms | |
#109 | Gemma 4 12B | 11.95B | INT4 Q8 FP16 | 105 tok/s 73 tok/s 43 tok/s | ~491 ms ~495 ms ~504 ms | |
#113 | Qwen3.5-9B | 9B | INT4 Q8 FP16 | 175 tok/s 111 tok/s 62 tok/s | ~370 ms ~373 ms ~381 ms | |
#114 | Qwen3.5-35B-A3B | 35B (3B active) | INT4 Q8 | 289 tok/s 210 tok/s | ~175 ms ~190 ms | |
#118 | K2 Horizon 7B | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
#128 | NVIDIA Nemotron 3 Nano 30B-A3B | 3.5B (30B active) | INT4 Q8 FP16 | 63 tok/s 37 tok/s 20 tok/s | ~1.2s ~1.2s ~1.2s | |
#129 | Gemma 4 31B | 30.7B | INT4 Q8 | 64 tok/s 36 tok/s | ~1.2s ~1.3s | |
#131 | Nemotron 3.5 Lightning | 30B (3B active) | INT4 Q8 | 129 tok/s 113 tok/s | ~175 ms ~191 ms | |
#132 | Qwen3.5-4B | 4B | INT4 Q8 FP16 | 292 tok/s 205 tok/s 124 tok/s | ~171 ms ~173 ms ~176 ms | |
#133 | Qwen3-30B-A3B | 30B (3B active) | INT4 Q8 | 292 tok/s 212 tok/s | ~171 ms ~187 ms | |
#136 | GPT-OSS 20B | 21B (3.6B active) | INT4 Q8 | 143 tok/s 125 tok/s | ~187 ms ~188 ms | |
#146 | Granite 4.2 8B | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#150 | Gemma 4 E4B | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#151 | Nemotron 3.5 Lightning 30B A3B | 30B (3B active) | INT4 Q8 | 292 tok/s 212 tok/s | ~171 ms ~187 ms | |
#153 | Qwen3-32B | 32B | INT4 Q8 | 62 tok/s 34 tok/s | ~1.3s ~1.4s | |
#154 | Qwen2.5-72B | 72B | INT4 | 26 tok/s | ~3.3s | |
#155 | Mistral Large 3 | 41B (675B active) | INT4 | 3 tok/s | ~28.0s | |
#156 | GLM-4 | 32B | INT4 Q8 | 49 tok/s 28 tok/s | ~1.3s ~1.5s | |
#157 | DeepSeek-R1 70B | 70B | INT4 | 29 tok/s | ~3.0s | |
#158 | GLM-4V | 9B | INT4 Q8 FP16 | 131 tok/s 93 tok/s 56 tok/s | ~372 ms ~375 ms ~382 ms | |
#159 | Gemma 3 27B | 27B | INT4 Q8 | 72 tok/s 42 tok/s | ~1.1s ~1.1s | |
#160 | Phi-4 | 14B | INT4 Q8 FP16 | 125 tok/s 76 tok/s 41 tok/s | ~570 ms ~575 ms ~586 ms | |
#161 | Llama 3.3 70B | 70B | INT4 | 29 tok/s | ~3.0s | |
#162 | DeepSeek-R1 32B | 32B | INT4 Q8 | 64 tok/s 35 tok/s | ~1.3s ~1.4s | |
#163 | Hunyuan Standard | 52B (389B active) | INT4 | 5 tok/s | ~16.2s | |
#164 | Qwen2.5-32B | 32B | INT4 Q8 | 62 tok/s 34 tok/s | ~1.3s ~1.4s | |
#168 | Magistral Small | 24B | INT4 Q8 | 61 tok/s 40 tok/s | ~967 ms ~976 ms | |
#169 | OLMo 3 7B Think | 7B | INT4 Q8 FP16 | 155 tok/s 113 tok/s 69 tok/s | ~292 ms ~294 ms ~300 ms | |
#170 | OLMo 3.1 32B Instruct | 32B | INT4 Q8 | 49 tok/s 28 tok/s | ~1.3s ~1.5s | |
#171 | Llama 3.1 70B | 70B | INT4 | 29 tok/s | ~3.0s | |
#172 | DeepSeek-R1 14B | 14B | INT4 Q8 FP16 | 128 tok/s 77 tok/s 41 tok/s | ~569 ms ~575 ms ~586 ms | |
#173 | Ministral 3 14B | 14B | INT4 Q8 FP16 | 93 tok/s 64 tok/s 35 tok/s | ~572 ms ~577 ms ~639 ms | |
#177 | Gemma 3 12B | 12B | INT4 Q8 FP16 | 141 tok/s 87 tok/s 47 tok/s | ~490 ms ~494 ms ~504 ms | |
#179 | Qwen2-72B | 72B | INT4 | 26 tok/s | ~3.3s | |
#180 | OLMo 3 32B Base | 32B | INT4 Q8 | 49 tok/s 28 tok/s | ~1.3s ~1.5s | |
#181 | Mistral-Small-2501 | 24B | INT4 Q8 | 80 tok/s 47 tok/s | ~963 ms ~972 ms | |
#182 | Ministral 3 8B | 8B | INT4 Q8 FP16 | 143 tok/s 102 tok/s 62 tok/s | ~332 ms ~334 ms ~341 ms | |
#183 | Qwen3.5-2B | 2B | INT4 Q8 FP16 | 397 tok/s 308 tok/s 206 tok/s | ~92 ms ~93 ms ~95 ms | |
#184 | MiniCPM5-2B | 2B | INT4 Q8 FP16 | 397 tok/s 308 tok/s 206 tok/s | ~92 ms ~93 ms ~95 ms | |
#185 | Qwen2.5-14B | 14B | INT4 Q8 FP16 | 125 tok/s 76 tok/s 41 tok/s | ~570 ms ~575 ms ~586 ms | |
#186 | Gemma 2 27B | 27B | INT4 Q8 | 72 tok/s 42 tok/s | ~1.1s ~1.1s | |
#187 | OLMo 3.1 32B Think | 32B | INT4 Q8 | 49 tok/s 28 tok/s | ~1.3s ~1.5s | |
#188 | Gemma 4 E2B | 5.1B | INT4 Q8 FP16 | 255 tok/s 173 tok/s 101 tok/s | ~215 ms ~217 ms ~221 ms | |
#189 | Qwen3-4B | 4B | INT4 Q8 FP16 | 292 tok/s 205 tok/s 124 tok/s | ~171 ms ~173 ms ~176 ms | |
#190 | DeepSeek-R1 8B | 8B | INT4 Q8 FP16 | 195 tok/s 124 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#191 | Llama 3 70B | 70B | INT4 | 29 tok/s | ~3.0s | |
#193 | Ministral 3 3B | 3B | INT4 Q8 FP16 | 244 tok/s 196 tok/s 133 tok/s | ~133 ms ~134 ms ~136 ms | |
#195 | Qwen3-14B | 14B | INT4 Q8 FP16 | 125 tok/s 76 tok/s 41 tok/s | ~570 ms ~575 ms ~586 ms | |
#196 | Kimi Linear 48B A3B Instruct | 48B (3B active) | INT4 | 106 tok/s | ~206 ms | |
#197 | Gemma 2 9B | 9B | INT4 Q8 FP16 | 175 tok/s 111 tok/s 62 tok/s | ~370 ms ~373 ms ~381 ms | |
#198 | Qwen2.5-7B | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
#199 | Qwen3-8B | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#200 | Gemma 3 4B | 4B | INT4 Q8 FP16 | 292 tok/s 205 tok/s 124 tok/s | ~171 ms ~173 ms ~176 ms | |
#201 | DeepSeek-R1 7B | 7B | INT4 Q8 FP16 | 213 tok/s 138 tok/s 78 tok/s | ~290 ms ~293 ms ~298 ms | |
#202 | Llama 3.1 8B | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#204 | Qwen3-1.7B | 1.7B | INT4 Q8 FP16 | 422 tok/s 335 tok/s 230 tok/s | ~79 ms ~80 ms ~81 ms | |
#205 | Command R | 35B | INT4 Q8 | 45 tok/s 26 tok/s | ~1.4s ~1.7s | |
#206 | Gemma 3 270M | 0.27B | INT4 Q8 FP16 | 576 tok/s 547 tok/s 489 tok/s | ~15 ms ~15 ms ~15 ms | |
#208 | Llama 3 8B | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#209 | Qwen3-0.6B | 0.6B | INT4 Q8 FP16 | 555 tok/s 495 tok/s 400 tok/s | ~30 ms ~30 ms ~31 ms | |
#210 | Phi-4-Mini | 3.8B | INT4 Q8 FP16 | 300 tok/s 212 tok/s 129 tok/s | ~163 ms ~165 ms ~168 ms | |
#211 | Mixtral-8x7B-v0.1 | 46.7B (7B active) | INT4 | 188 tok/s | ~336 ms | |
#212 | Ministral-8B-2410 | 8B | INT4 Q8 FP16 | 191 tok/s 123 tok/s 69 tok/s | ~330 ms ~333 ms ~339 ms | |
#213 | DeepSeek-R1 1.5B | 1.5B | INT4 Q8 FP16 | 440 tok/s 356 tok/s 249 tok/s | ~71 ms ~71 ms ~72 ms | |
#214 | Llama 3.2 3B | 3B | INT4 Q8 FP16 | 336 tok/s 246 tok/s 155 tok/s | ~132 ms ~133 ms ~135 ms | |
#215 | Qwen3.5-0.8B | 0.8B | INT4 Q8 FP16 | 523 tok/s 454 tok/s 352 tok/s | ~39 ms ~40 ms ~40 ms | |
#216 | Phi-3-mini | 3.8B | INT4 Q8 FP16 | 300 tok/s 212 tok/s 129 tok/s | ~163 ms ~165 ms ~168 ms | |
#217 | Gemma 3n E2B IT | 6B (2B active) | INT4 Q8 FP16 | 244 tok/s 213 tok/s 159 tok/s | ~100 ms ~100 ms ~102 ms | |
#218 | Mistral-7B-Instruct-v0.1 | 7.3B | INT4 Q8 FP16 | 203 tok/s 132 tok/s 74 tok/s | ~302 ms ~305 ms ~310 ms | |
#219 | Llama 3.2 1B | 1B | INT4 Q8 FP16 | 494 tok/s 419 tok/s 314 tok/s | ~49 ms ~50 ms ~50 ms | |
#220 | Gemma 2 2B | 2B | INT4 Q8 FP16 | 397 tok/s 308 tok/s 206 tok/s | ~92 ms ~93 ms ~95 ms | |
#221 | Phi-3-medium | 14B | INT4 Q8 FP16 | 125 tok/s 76 tok/s 41 tok/s | ~570 ms ~575 ms ~586 ms | |
#222 | OLMo 3 7B Instruct | 7B | INT4 Q8 FP16 | 155 tok/s 113 tok/s 69 tok/s | ~292 ms ~294 ms ~300 ms | |
#223 | Yi-34B | 34B | INT4 Q8 | 46 tok/s 27 tok/s | ~1.4s ~1.6s | |
#224 | Qwen2-7B | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
#225 | Phi-3-small | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
#226 | Gemma 3 1B | 1B | INT4 Q8 FP16 | 494 tok/s 419 tok/s 314 tok/s | ~49 ms ~50 ms ~50 ms | |
#227 | Mistral-7B-Instruct-v0.2 | 7.3B | INT4 Q8 FP16 | 203 tok/s 132 tok/s 74 tok/s | ~302 ms ~305 ms ~310 ms | |
#229 | Gemma 1 7B | 7B | INT4 Q8 FP16 | 155 tok/s 113 tok/s 69 tok/s | ~292 ms ~294 ms ~300 ms | |
#230 | Gemma 1 2B | 2B | INT4 Q8 FP16 | 415 tok/s 318 tok/s 211 tok/s | ~92 ms ~93 ms ~95 ms | |
#231 | ChatGLM3-6B | 6B | INT4 Q8 FP16 | 171 tok/s 126 tok/s 79 tok/s | ~252 ms ~254 ms ~259 ms | |
#232 | ChatGLM2-6B | 6B | INT4 Q8 FP16 | 171 tok/s 126 tok/s 79 tok/s | ~252 ms ~254 ms ~259 ms | |
#233 | ChatGLM-6B | 6B | INT4 Q8 FP16 | 171 tok/s 126 tok/s 79 tok/s | ~252 ms ~254 ms ~259 ms | |
Unranked Models (Sorted by Speed) | ||||||
- | Qwen2-0.5B | 0.5B | INT4 Q8 FP16 | 571 tok/s 518 tok/s 429 tok/s | ~25 ms ~25 ms ~26 ms | |
- | Qwen2.5-0.5B | 0.5B | INT4 Q8 FP16 | 571 tok/s 518 tok/s 429 tok/s | ~25 ms ~25 ms ~26 ms | |
- | ERNIE-4.5-0.3B | 0.3B | INT4 Q8 FP16 | 565 tok/s 535 tok/s 474 tok/s | ~17 ms ~17 ms ~17 ms | |
- | ERNIE-4.5-0.3B-Base | 0.3B | INT4 Q8 FP16 | 565 tok/s 535 tok/s 474 tok/s | ~17 ms ~17 ms ~17 ms | |
- | Falcon-1B | 1B | INT4 Q8 FP16 | 511 tok/s 430 tok/s 320 tok/s | ~49 ms ~49 ms ~50 ms | |
- | Falcon3-1B | 1B | INT4 Q8 FP16 | 494 tok/s 419 tok/s 314 tok/s | ~49 ms ~50 ms ~50 ms | |
- | Qwen2-1.5B | 1.5B | INT4 Q8 FP16 | 440 tok/s 356 tok/s 249 tok/s | ~71 ms ~71 ms ~72 ms | |
- | Qwen2.5-1.5B | 1.5B | INT4 Q8 FP16 | 440 tok/s 356 tok/s 249 tok/s | ~71 ms ~71 ms ~72 ms | |
- | Spark X2.5 1.7B | 1.7B | INT4 Q8 FP16 | 422 tok/s 335 tok/s 230 tok/s | ~79 ms ~80 ms ~81 ms | |
- | Hy-MT2 1.8B | 1.8B | INT4 Q8 FP16 | 414 tok/s 326 tok/s 221 tok/s | ~83 ms ~84 ms ~86 ms | |
- | Falcon-3B | 3B | INT4 Q8 FP16 | 351 tok/s 253 tok/s 157 tok/s | ~131 ms ~133 ms ~135 ms | |
- | CroissantLLM Base | 1.3B | INT4 Q8 FP16 | 346 tok/s 303 tok/s 230 tok/s | ~63 ms ~63 ms ~64 ms | |
- | Phi-1 | 1.3B | INT4 Q8 FP16 | 346 tok/s 303 tok/s 230 tok/s | ~63 ms ~63 ms ~64 ms | |
- | Phi-1.5 | 1.3B | INT4 Q8 FP16 | 346 tok/s 303 tok/s 230 tok/s | ~63 ms ~63 ms ~64 ms | |
- | DeepSeek-R1 3B | 3B | INT4 Q8 FP16 | 340 tok/s 248 tok/s 155 tok/s | ~132 ms ~133 ms ~135 ms | |
- | Falcon3-3B | 3B | INT4 Q8 FP16 | 336 tok/s 246 tok/s 155 tok/s | ~132 ms ~133 ms ~135 ms | |
- | Qwen2.5-3B | 3B | INT4 Q8 FP16 | 336 tok/s 246 tok/s 155 tok/s | ~132 ms ~133 ms ~135 ms | |
- | ERNIE-4.5-21B-A3B | 21B (3B active) | INT4 Q8 | 303 tok/s 229 tok/s | ~160 ms ~161 ms | |
- | ERNIE-4.5-21B-A3B-Base | 21B (3B active) | INT4 Q8 | 303 tok/s 229 tok/s | ~160 ms ~161 ms | |
- | ERNIE-4.5-VL-28B-A3B | 28B (3B active) | INT4 Q8 | 294 tok/s 225 tok/s | ~169 ms ~170 ms | |
- | ERNIE-4.5-VL-28B-A3B-Base | 28B (3B active) | INT4 Q8 | 294 tok/s 225 tok/s | ~169 ms ~170 ms | |
- | Hy-MT2 30B A3B | 30B (3B active) | INT4 Q8 | 292 tok/s 212 tok/s | ~171 ms ~187 ms | |
- | Spark X2.5 4B | 4B | INT4 Q8 FP16 | 292 tok/s 205 tok/s 124 tok/s | ~171 ms ~173 ms ~176 ms | |
- | Phi-2 | 2.7B | INT4 Q8 FP16 | 254 tok/s 207 tok/s 143 tok/s | ~121 ms ~122 ms ~124 ms | |
- | MaLLaM-3B | 3B | INT4 Q8 FP16 | 244 tok/s 196 tok/s 133 tok/s | ~133 ms ~134 ms ~136 ms | |
- | SmolLM3 3B | 3B | INT4 Q8 FP16 | 244 tok/s 196 tok/s 133 tok/s | ~133 ms ~134 ms ~136 ms | |
- | Falcon-7B | 7B | INT4 Q8 FP16 | 217 tok/s 139 tok/s 78 tok/s | ~290 ms ~293 ms ~298 ms | |
- | Falcon3-7B | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
- | Hy-MT2 7B | 7B | INT4 Q8 FP16 | 209 tok/s 136 tok/s 77 tok/s | ~290 ms ~293 ms ~298 ms | |
- | Mistral-7B-v0.1 | 7.3B | INT4 Q8 FP16 | 203 tok/s 132 tok/s 74 tok/s | ~302 ms ~305 ms ~310 ms | |
- | LensVLM 9B | 9B | INT4 Q8 FP16 | 175 tok/s 111 tok/s 62 tok/s | ~370 ms ~373 ms ~381 ms | |
- | MiMo V2.6 Distill Qwen 9B | 9B | INT4 Q8 FP16 | 175 tok/s 111 tok/s 62 tok/s | ~370 ms ~373 ms ~381 ms | |
- | ChatGLM3-6B-32K | 6B | INT4 Q8 FP16 | 171 tok/s 126 tok/s 79 tok/s | ~252 ms ~254 ms ~259 ms | |
- | Yi-6B | 6B | INT4 Q8 FP16 | 171 tok/s 126 tok/s 79 tok/s | ~252 ms ~254 ms ~259 ms | |
- | Kimi-VL-A3B-Instruct | 16B (3B active) | INT4 Q8 FP16 | 163 tok/s 143 tok/s 101 tok/s | ~157 ms ~158 ms ~174 ms | |
- | Falcon3-10B | 10B | INT4 Q8 FP16 | 162 tok/s 102 tok/s 56 tok/s | ~410 ms ~413 ms ~421 ms | |
- | Falcon2-11B | 11B | INT4 Q8 FP16 | 157 tok/s 96 tok/s 52 tok/s | ~449 ms ~453 ms ~462 ms | |
- | Hunyuan Lite | 7B | INT4 Q8 FP16 | 155 tok/s 113 tok/s 69 tok/s | ~292 ms ~294 ms ~300 ms | |
- | OLMo 3 7B Base | 7B | INT4 Q8 FP16 | 155 tok/s 113 tok/s 69 tok/s | ~292 ms ~294 ms ~300 ms | |
- | SEA-LION-7B | 7.1B | INT4 Q8 FP16 | 154 tok/s 112 tok/s 68 tok/s | ~296 ms ~298 ms ~304 ms | |
- | SEA-LION-7B-Instruct | 7.1B | INT4 Q8 FP16 | 154 tok/s 112 tok/s 68 tok/s | ~296 ms ~298 ms ~304 ms | |
- | Sahabat-AI-Llama3-8B-Instruct | 8B | INT4 Q8 FP16 | 143 tok/s 102 tok/s 62 tok/s | ~332 ms ~334 ms ~341 ms | |
- | Typhoon-2-8B | 8B | INT4 Q8 FP16 | 143 tok/s 102 tok/s 62 tok/s | ~332 ms ~334 ms ~341 ms | |
- | GLM-4-9B | 9B | INT4 Q8 FP16 | 131 tok/s 93 tok/s 56 tok/s | ~372 ms ~375 ms ~382 ms | |
- | GLM-4-9B-Chat | 9B | INT4 Q8 FP16 | 131 tok/s 93 tok/s 56 tok/s | ~372 ms ~375 ms ~382 ms | |
- | GLM-4-9B-Chat-1M | 9B | INT4 Q8 FP16 | 131 tok/s 93 tok/s 56 tok/s | ~372 ms ~375 ms ~382 ms | |
- | Yi-9B | 9B | INT4 Q8 FP16 | 131 tok/s 93 tok/s 56 tok/s | ~372 ms ~375 ms ~382 ms | |
- | Sahabat-AI-Gemma2-9B | 9.2B | INT4 Q8 FP16 | 129 tok/s 91 tok/s 54 tok/s | ~380 ms ~383 ms ~390 ms | |
- | Sahabat-AI-Gemma2-9B-Instruct | 9.2B | INT4 Q8 FP16 | 129 tok/s 91 tok/s 54 tok/s | ~380 ms ~383 ms ~390 ms | |
- | GreenMind-14B-R1 | 14B | INT4 Q8 FP16 | 93 tok/s 64 tok/s 35 tok/s | ~572 ms ~577 ms ~639 ms | |
- | Falcon-40B | 40B | INT4 Q8 | 52 tok/s 27 tok/s | ~1.6s ~1.9s | |
Inference Capacity
Maximum model parameter class runnable by weight precision.
Context:
4-bit
32B Models
8-bit
32B Models
16-bit
14B 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 A100 (40GB).
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 1555 GB/s aggregate bandwidth, batch size 1 inference operates in a memory-bandwidth bound regime. Generates approximately 346 tok/s on an 8B Q4 model and 41 tok/s on a 70B Q4 model.
Frequently Asked Questions
©2025 ApX Machine Learning