Benchmarks
Real performance data, not marketing claims. Our benchmarks test every GPU we offer across LLM inference, image generation, OCR, and TTS workloads on dedicated GPU servers. See our tokens/sec benchmark for the latest results.
Gemma 2 9B on RTX 4060 Ti: Performance Benchmark & Cost, Category: Benchmarks, Slug: gemma-2-9b-on-rtx-4060-ti-benchmark, Excerpt: Gemma 2 9B benchmarked on RTX 4060 Ti: 23.6 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Gemma 2 9B benchmarked on RTX 4060 Ti: 23.6 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 -->
Gemma 2 9B on RTX 3090: Performance Benchmark & Cost, Category: Benchmarks, Slug: gemma-2-9b-on-rtx-3090-benchmark, Excerpt: Gemma 2 9B benchmarked on RTX 3090: 52.0 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Gemma 2 9B benchmarked on RTX 3090: 52.0 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration.,…
Phi-3 Mini on RTX 5080: Performance Benchmark & Cost, Category: Benchmarks, Slug: phi-3-mini-on-rtx-5080-benchmark, Excerpt: Phi-3 Mini benchmarked on RTX 5080: 82 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Phi-3 Mini benchmarked on RTX 5080: 82 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal…
Phi-3 Mini on RTX 5090: Performance Benchmark & Cost, Category: Benchmarks, Slug: phi-3-mini-on-rtx-5090-benchmark, Excerpt: Phi-3 Mini benchmarked on RTX 5090: 100 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Phi-3 Mini benchmarked on RTX 5090: 100 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal…
Gemma 2 9B on RTX 5080: Performance Benchmark & Cost, Category: Benchmarks, Slug: gemma-2-9b-on-rtx-5080-benchmark, Excerpt: Gemma 2 9B benchmarked on RTX 5080: 48.8 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Gemma 2 9B benchmarked on RTX 5080: 48.8 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and…
Gemma 2 9B on RTX 5090: Performance Benchmark & Cost, Category: Benchmarks, Slug: gemma-2-9b-on-rtx-5090-benchmark, Excerpt: Gemma 2 9B benchmarked on RTX 5090: 112.3 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Gemma 2 9B benchmarked on RTX 5090: 112.3 tok/s at FP16, VRAM usage, cost per 1M tokens, and deployment configuration.,…
LLaMA 3 70B on RTX 3090: Performance Benchmark & Cost, Category: Benchmarks, Slug: llama-3-70b-on-rtx-3090-benchmark, Excerpt: LLaMA 3 70B benchmarked on RTX 3090: 5.2 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
LLaMA 3 70B benchmarked on RTX 3090: 5.2 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and…
LLaMA 3 70B on RTX 5090: Performance Benchmark & Cost, Category: Benchmarks, Slug: llama-3-70b-on-rtx-5090-benchmark, Excerpt: LLaMA 3 70B benchmarked on RTX 5090: 12.8 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
LLaMA 3 70B benchmarked on RTX 5090: 12.8 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and…
Mixtral 8x7B on RTX 3090: Performance Benchmark & Cost, Category: Benchmarks, Slug: mixtral-8x7b-on-rtx-3090-benchmark, Excerpt: Mixtral 8x7B benchmarked on RTX 3090: 18 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Mixtral 8x7B benchmarked on RTX 3090: 18 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment…
Mixtral 8x7B on RTX 5080: Performance Benchmark & Cost, Category: Benchmarks, Slug: mixtral-8x7b-on-rtx-5080-benchmark, Excerpt: Mixtral 8x7B benchmarked on RTX 5080: 32 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment configuration., Internal links: 9 –>
Mixtral 8x7B benchmarked on RTX 5080: 32 tok/s at 4-bit GGUF Q4_K_M, VRAM usage, cost per 1M tokens, and deployment…
Explore GPU Hosting Solutions
From the blog to your next deployment — pick the right platform for your workload.
Tokens/sec Benchmarks
Real-world tokens per second data across every GPU we offer, tested on popular LLMs.
View BenchmarksTTS Latency Benchmarks
Time-to-first-audio for Coqui, Bark, Kokoro, and XTTS-v2 across GPU tiers.
View TTS BenchmarksOCR Speed Benchmarks
Pages per second for PaddleOCR and Tesseract across our GPU server lineup.
View OCR BenchmarksCost per 1M Tokens
What does it cost to process a million tokens on each GPU? Interactive calculator.
Calculate CostDedicated GPU Hosting
Bare-metal servers with a dedicated GPU, NVMe, full root access, and 1Gbps networking from our UK datacenter.
Browse GPU ServersOpen Source LLM Hosting
Deploy LLaMA, Mistral, DeepSeek, and more on dedicated hardware with no per-token API fees.
Explore LLM HostingReady to deploy your AI workload?
Dedicated GPU servers from our UK datacenter. NVMe storage, 1Gbps networking, full root access.
65 Victory Avenue
PE7 1XU, England
GPUAIR © All Rights Reserved.