Compute Matrix Directory (Page 1)

💻 DeepSeek-V3 671B MoE (FP16 Uncompressed Native) on NVIDIA H100 80GB SXM5 VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in FP16 Uncompressed Native deployed on NVIDIA H100 80GB SXM5.

💻 DeepSeek-V3 671B MoE (BF16 Bfloat16 Mixed Precision) on NVIDIA H200 141GB HBM3e VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in BF16 Bfloat16 Mixed Precision deployed on NVIDIA H200 141GB HBM3e.

💻 DeepSeek-V3 671B MoE (FP8 Scaled Native Hopper) on NVIDIA B200 192GB Blackwell VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in FP8 Scaled Native Hopper deployed on NVIDIA B200 192GB Blackwell.

💻 DeepSeek-V3 671B MoE (INT8 SmoothQuant Precision) on NVIDIA A100 80GB PCIe VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in INT8 SmoothQuant Precision deployed on NVIDIA A100 80GB PCIe.

💻 DeepSeek-V3 671B MoE (AWQ 4-Bit Activation-Aware) on NVIDIA RTX 4090 24GB GDDR6X VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in AWQ 4-Bit Activation-Aware deployed on NVIDIA RTX 4090 24GB GDDR6X.

💻 DeepSeek-V3 671B MoE (GPTQ 4-Bit Second-Order) on NVIDIA L40S 48GB Ada Lovelace VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in GPTQ 4-Bit Second-Order deployed on NVIDIA L40S 48GB Ada Lovelace.

💻 DeepSeek-V3 671B MoE (GGUF Q4_K_M Medium Quant) on AMD Instinct MI300X 192GB VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-V3 671B MoE quantized in GGUF Q4_K_M Medium Quant deployed on AMD Instinct MI300X 192GB.

💻 DeepSeek-R1 Reasoning 671B (FP16 Uncompressed Native) on NVIDIA H100 80GB SXM5 VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in FP16 Uncompressed Native deployed on NVIDIA H100 80GB SXM5.

💻 DeepSeek-R1 Reasoning 671B (BF16 Bfloat16 Mixed Precision) on NVIDIA H200 141GB HBM3e VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in BF16 Bfloat16 Mixed Precision deployed on NVIDIA H200 141GB HBM3e.

💻 DeepSeek-R1 Reasoning 671B (FP8 Scaled Native Hopper) on NVIDIA B200 192GB Blackwell VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in FP8 Scaled Native Hopper deployed on NVIDIA B200 192GB Blackwell.

💻 DeepSeek-R1 Reasoning 671B (INT8 SmoothQuant Precision) on NVIDIA A100 80GB PCIe VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in INT8 SmoothQuant Precision deployed on NVIDIA A100 80GB PCIe.

💻 DeepSeek-R1 Reasoning 671B (AWQ 4-Bit Activation-Aware) on NVIDIA RTX 4090 24GB GDDR6X VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in AWQ 4-Bit Activation-Aware deployed on NVIDIA RTX 4090 24GB GDDR6X.

💻 DeepSeek-R1 Reasoning 671B (GPTQ 4-Bit Second-Order) on NVIDIA L40S 48GB Ada Lovelace VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in GPTQ 4-Bit Second-Order deployed on NVIDIA L40S 48GB Ada Lovelace.

💻 DeepSeek-R1 Reasoning 671B (GGUF Q4_K_M Medium Quant) on AMD Instinct MI300X 192GB VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for DeepSeek-R1 Reasoning 671B quantized in GGUF Q4_K_M Medium Quant deployed on AMD Instinct MI300X 192GB.

💻 Llama-3.1 405B Ultra-Scale (FP16 Uncompressed Native) on NVIDIA H100 80GB SXM5 VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in FP16 Uncompressed Native deployed on NVIDIA H100 80GB SXM5.

💻 Llama-3.1 405B Ultra-Scale (BF16 Bfloat16 Mixed Precision) on NVIDIA H200 141GB HBM3e VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in BF16 Bfloat16 Mixed Precision deployed on NVIDIA H200 141GB HBM3e.

💻 Llama-3.1 405B Ultra-Scale (FP8 Scaled Native Hopper) on NVIDIA B200 192GB Blackwell VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in FP8 Scaled Native Hopper deployed on NVIDIA B200 192GB Blackwell.

💻 Llama-3.1 405B Ultra-Scale (INT8 SmoothQuant Precision) on NVIDIA A100 80GB PCIe VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in INT8 SmoothQuant Precision deployed on NVIDIA A100 80GB PCIe.

💻 Llama-3.1 405B Ultra-Scale (AWQ 4-Bit Activation-Aware) on NVIDIA RTX 4090 24GB GDDR6X VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in AWQ 4-Bit Activation-Aware deployed on NVIDIA RTX 4090 24GB GDDR6X.

💻 Llama-3.1 405B Ultra-Scale (GPTQ 4-Bit Second-Order) on NVIDIA L40S 48GB Ada Lovelace VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in GPTQ 4-Bit Second-Order deployed on NVIDIA L40S 48GB Ada Lovelace.

💻 Llama-3.1 405B Ultra-Scale (GGUF Q4_K_M Medium Quant) on AMD Instinct MI300X 192GB VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 405B Ultra-Scale quantized in GGUF Q4_K_M Medium Quant deployed on AMD Instinct MI300X 192GB.

💻 Llama-3.3 70B High-Efficiency (FP16 Uncompressed Native) on NVIDIA H100 80GB SXM5 VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in FP16 Uncompressed Native deployed on NVIDIA H100 80GB SXM5.

💻 Llama-3.3 70B High-Efficiency (BF16 Bfloat16 Mixed Precision) on NVIDIA H200 141GB HBM3e VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in BF16 Bfloat16 Mixed Precision deployed on NVIDIA H200 141GB HBM3e.

💻 Llama-3.3 70B High-Efficiency (FP8 Scaled Native Hopper) on NVIDIA B200 192GB Blackwell VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in FP8 Scaled Native Hopper deployed on NVIDIA B200 192GB Blackwell.

💻 Llama-3.3 70B High-Efficiency (INT8 SmoothQuant Precision) on NVIDIA A100 80GB PCIe VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in INT8 SmoothQuant Precision deployed on NVIDIA A100 80GB PCIe.

💻 Llama-3.3 70B High-Efficiency (AWQ 4-Bit Activation-Aware) on NVIDIA RTX 4090 24GB GDDR6X VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in AWQ 4-Bit Activation-Aware deployed on NVIDIA RTX 4090 24GB GDDR6X.

💻 Llama-3.3 70B High-Efficiency (GPTQ 4-Bit Second-Order) on NVIDIA L40S 48GB Ada Lovelace VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in GPTQ 4-Bit Second-Order deployed on NVIDIA L40S 48GB Ada Lovelace.

💻 Llama-3.3 70B High-Efficiency (GGUF Q4_K_M Medium Quant) on AMD Instinct MI300X 192GB VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.3 70B High-Efficiency quantized in GGUF Q4_K_M Medium Quant deployed on AMD Instinct MI300X 192GB.

💻 Llama-3.1 70B Enterprise (FP16 Uncompressed Native) on NVIDIA H100 80GB SXM5 VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 70B Enterprise quantized in FP16 Uncompressed Native deployed on NVIDIA H100 80GB SXM5.

💻 Llama-3.1 70B Enterprise (BF16 Bfloat16 Mixed Precision) on NVIDIA H200 141GB HBM3e VRAM & Throughput Calculator

Exact VRAM memory allocation, dynamic KV-cache requirements, and tensor parallelism slicing for Llama-3.1 70B Enterprise quantized in BF16 Bfloat16 Mixed Precision deployed on NVIDIA H200 141GB HBM3e.

High-Performance Compute Tools Directory