NVIDIA
Operating mode
Use this to bias workload recommendations toward responsiveness, background autonomy, lighter serving, or multi-GPU scale-out.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
The NVIDIA H100 PCIe is the server-accessible variant of the H100 flagship, delivering 80 GB of HBM3 and full FP8 Transformer Engine support in a standard PCIe 5.0 form factor. Compared to the H100 SXM, it trades some bandwidth (2.0 TB/s vs. 3.35 TB/s) and compute (756 TFLOPS vs. 989 TFLOPS FP16) for compatibility with standard servers that lack SXM5 baseboard infrastructure. It remains a very capable inference GPU — able to run 70B models at FP16 and 4x faster than an A100 for LLM inference tasks. For teams that cannot afford SXM infrastructure, the H100 PCIe offers the Hopper Transformer Engine advantage in a drop-in form.
Beyond LLMs
What AI tasks this GPU can handle — from text generation to image and video creation.
| Capability | Status | Representative Model |
|---|---|---|
| LLM Chat (7B) | Runs natively | Llama 3.1 8B Q4 |
| LLM Coding (30B) | Runs natively | Qwen 3 30B Q4 |
| LLM Large (70B) | Runs natively | Llama 3.1 70B Q4 |
| Image Gen (SDXL) | Runs natively | SDXL 1.0 FP16 |
| Image Gen (Flux) | Runs natively | Flux.1 Dev FP16 |
| Image Gen (SD 3.5) | Runs natively | SD 3.5 Large FP16 |
| Video Short (25f) | Runs natively | LTX Video 2B |
| Video Long (100f) | Runs with offload | Wan Video 14B |
Architecture
Hopper is NVIDIA's datacenter-focused architecture succeeding Ampere. Built on TSMC 4N, it introduces the Transformer Engine with automatic FP8/FP16 mixed-precision training, HBM3/HBM3e memory, and NVLink 4.0 for multi-GPU scaling. The H100 flagship delivers up to 3x the AI training performance of A100.
AI Relevance
The Transformer Engine automatically manages FP8 precision for optimal training speed without accuracy loss. With up to 141 GB HBM3e (H200), Hopper GPUs can hold the largest open-weight models entirely in GPU memory, making them the workhorse of AI datacenters.
Conselho de compra
Excelente escolha para IA local
Roda 36 de 50 modelos principais bem — um ótimo coringa para inferência local.
80.0 GB
VRAM
$30,000
Preço sugerido
$375/GB
Custo por GB de VRAM
Melhores modelos para esta GPU
What will limit you first
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
Best upgrade itinerary
Unlocks 1 additional models that do not fit on the current setup.
Quer mais margem? Mac Studio M2 Ultra 128GB (128.0 GB unified memory) é o próximo passo.
Chat
SThis model is a direct match for chat. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Coding
SThis model is a direct match for coding. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Agentic Coding
SThis model is still usable for agentic-coding, but it is not the most specialized pick. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Reasoning
SThis model is a direct match for reasoning. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
RAG
SThis model is a direct match for rag. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Qwen3-Coder-NextQwen 2.5 VL 72BQwen 3.6 35B A3BQwen3-Coder 30B A3B InstructQwen 3.5 27BQwen 3.5 35B A3BQwen3-VL 30B A3B InstructQwen 3 32BQwen 3.6 27B
Magistral Small 2507Devstral Small 2 24B InstructQwen 3 30B A3BNemotron 3 Nano 30B
Command A 111B
Gemma 4 31BDevstral Small 1.1Qwen 3.5 9BQwen 3 14BNemotron Cascade 2 30B A3B
Phi-4-reasoning-plus 14B
GPT-OSS 20BQwen 3 8B
EXAONE 4.0 32BGemma 4 26B A4BQwen 3.5 4BQwen 3.5 122B A10BMistral Small 4 119BMinistral 3 14BDevstral 2 123B InstructNemotron Nano 8BPhi-4 Mini Reasoning 4BGPT-OSS 120BPixtral Large 124BLeanstral 119B A6B
Jina Embeddings v3
BGE M3Qwen 3.5 397B A17B
Kimi K2.5Kimi K2.6
DeepSeek V4 ProDeepSeek V4 Flash
GLM-5.1GLM-5DeepSeek V3.2Qwen 3 235B A22BQwen3-Coder 480B A35B InstructDeepSeek Coder V2 236BDeepSeek R1 671BDeepSeek V3.1 671BQuase ao alcance
Modelos de alta qualidade que precisam de um pouco mais de memória
Image & Video Generation
51 of 52 models can generate images or video on your NVIDIA H100 PCIe 80GB
| Model | Max Resolution | Gen Time | Grade |
|---|---|---|---|
| SD TurboImage | 512×512 | 0ms | S |
| Stable Diffusion 1.5Image | 512×768 | 100ms | S |
| Realistic Vision v5.1Image | 512×768 | 100ms | S |
| DreamShaper 8Image | 512×768 | 100ms | S |
| LCM DreamShaper v7Image | 512×768 | 0ms | S |
| PixArt-SigmaImage | 1024×1024 | 400ms | S |
| FramePack I2VVideo | 1280×720 | 700ms/frame | S |
| SDXL TurboImage | 512×512 | 0ms | S |
| SDXL LightningImage | 1024×1024 | 100ms | S |
| Stable Diffusion XL 1.0Image | 1024×1024 | 400ms | S |
| Playground v2.5Image | 1024×1024 | 600ms | S |
| RealVisXL v5.0Image | 1024×1024 | 400ms | S |
| DreamShaper XLImage | 1024×1024 | 400ms | S |
| Juggernaut XL v9Image | 1024×1024 | 400ms | S |
| Animagine XL 3.1Image | 1024×1024 | 400ms | S |
| Pony Diffusion V6 XLImage | 1024×1024 | 400ms | S |
| Animagine XL 4.0Image | 1024×1024 | 400ms | S |
| Illustrious XLImage | 1024×1024 | 400ms | S |
| Wan Video 2.1 1.3BVideo | 480×832 | 300ms/frame | S |
| Stable Diffusion 3.5 MediumImage | 1024×1024 | 700ms | S |
| Flux.2 Klein 4BImage | 1024×1024 | 100ms | S |
| LTX Video 2BVideo | 1280×720 | 300ms/frame | S |
| KolorsImage | 1024×1024 | 800ms | S |
| Stable CascadeImage | 1024×1024 | ~1s | S |
| AuraFlow v0.3Image | 1536×1536 | ~1.8s | S |
| Stable Diffusion 3.5 LargeImage | 1024×1024 | ~2.2s | S |
| Stable Diffusion 3.5 Large TurboImage | 1024×1024 | 400ms | S |
| CogVideoX 2BVideo | 720×480 | 300ms/frame | S |
| HunyuanVideoVideo | 720×1280 | 700ms/frame | S |
| ChromaImage | 1024×1024 | 400ms | S |
| Z-Image TurboImage | 1536×1536 | 400ms | S |
| Flux.1 DevImage | 1024×1024 | ~1.8s | S |
| Flux.1 SchnellImage | 1024×1024 | 300ms | S |
| LTX Video 13BVideo | 1280×720 | 700ms/frame | S |
| Flux.1 Kontext DevImage | 1024×1024 | ~2s | S |
| AnimateDiff v1.5.3Video | 512×768 | 200ms/frame | S |
| Cosmos Diffusion 7BVideo | 1024×576 | 600ms/frame | S |
| CogVideoX 5BVideo | 720×480 | 500ms/frame | S |
| Wan2.2 TI2V 5BVideo | 832×480 | 500ms/frame | S |
| Flux.2 Klein 9BImage | 1024×1024 | 200ms | S |
| Flux.1 Fill DevImage | 1024×1024 | ~1.7s | S |
| Mochi 1 PreviewVideo | 848×480 | 600ms/frame | S |
| HunyuanVideo 1.5Video | 720×1280 | 600ms/frame | S |
| Helios 14BVideo | 1280×720 | 700ms/frame | S |
| SkyReels V2 14BVideo | 1280×720 | 700ms/frame | S |
| Wan Video 2.1 14BVideo | 720×1280 | 700ms/frame | S |
| Wan Video 2.2 14BVideo | 720×1280 | 700ms/frame | S |
| Qwen ImageImage | 1024×1024 | 700ms | S |
| Qwen Image EditImage | 1024×1024 | 700ms | S |
| Flux.2 DevImage | 1024×1024 | ~18.6s | S |
| MAGI-1Video | 1280×720 | 900ms/frame | A |
| HunyuanImage 3.0Image | 256×256 | ~1.2s | F |
Image models estimated at 1024×1024 (28 steps, FP16). Video models estimated at 768×512 (25 frames, 30 steps, FP16). Actual performance varies with runtime and system load.
Multi-GPU scaling
Scale out with multiple GPUs for larger models. PCIe interconnect with 22% scaling overhead.
| Config | Effective memory | Models that fit | Est. bandwidth |
|---|---|---|---|
| 1× NVIDIA | 80 GB | 350/374 | 2,000 GB/s |
| 2× NVIDIA | 160 GB | 359/374 | 3,120 GB/s |
| 4× NVIDIA | 320 GB | 364/374 | 6,240 GB/s |
Model counts use default quantization at coding workload settings. Multi-GPU scaling factor: 0.78× per additional GPU.
Upgrade paths
See what you unlock with more powerful hardware
Opções de upgrade
4× NVIDIA H100 PCIe 80GBMulti-GPUUnlocks 14 additional models that do not fit on the current setup.
Lifts average decode speed across fitting models by about 55%.
Scale-out only pays off if the host platform has enough PCIe lanes, slot spacing, power, and cooling.
The bigger the setup gets, the more the runtime matters. Multi-GPU and multi-user serving are where vLLM, SGLang, TGI, TensorRT-LLM, or tuned llama.cpp start to earn their complexity.
~$30,000 MSRP
Unlocks 1 additional models that do not fit on the current setup.
~$3,999 MSRP
RTX PRO 6000 Blackwell Server Edition 96GBUpgrade NVIDIAUnlocks 1 additional models that do not fit on the current setup.
~$9,999 MSRP
Unlocks 13 additional models that do not fit on the current setup.
Lifts average decode speed across fitting models by about 44%.
~$20,000 MSRP
Unlocks 14 additional models that do not fit on the current setup.
Lifts average decode speed across fitting models by about 61%.
~$8,000 MSRP
NVIDIA H100 PCIe 80GB (80 GB VRAM) can run these top models: Qwen3-Coder-Next (score: 97/100), Qwen 2.5 VL 72B (score: 95/100), Qwen 3.6 35B A3B (score: 93/100). See the full compatibility list above.
NVIDIA H100 PCIe 80GB has 80 GB of VRAM available for AI model inference. This determines which models and quantization levels you can run locally.
Yes, NVIDIA H100 PCIe 80GB is excellent for running LLMs locally with top compatibility scores above 80/100.
For coding on NVIDIA H100 PCIe 80GB, we recommend Qwen3-Coder-Next. It achieves 113.5 tokens per second with 244K context window. This model is a direct match for coding. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
There are 5 upgrade path(s) from NVIDIA H100 PCIe 80GB: NVIDIA H100 PCIe 80GB, Mac Studio M2 Ultra 128GB. Upgrading would unlock larger models and faster inference speeds.
Yes, NVIDIA H100 PCIe 80GB with 80 GB of usable memory can run Flux.1 Dev at FP16 natively. Flux is a 12B parameter diffusion transformer that produces high-quality images. You can also run the Schnell variant for faster generation.
NVIDIA H100 PCIe 80GB (80 GB VRAM) can handle various AI generation tasks beyond LLMs. For image generation, SDXL and Stable Diffusion 3.5 run well. Flux.1 Dev also runs natively for state-of-the-art image quality. For video, LTX Video 2.3 can generate short clips. Check the AI Capability Matrix above for detailed compatibility.
NVIDIA H100 PCIe 80GB is excellent for AI image generation. With 80 GB of usable memory, it runs all major diffusion models including Flux.1, SDXL, and Stable Diffusion 3.5 at full precision. You can generate high-resolution images quickly and even handle video generation models.
Yes, NVIDIA H100 PCIe 80GB with 80 GB of usable memory can run Qwen 3.5 27B at Q8 (near-lossless, ~28.9 GB) or even FP16 (~55.4 GB) depending on your context needs. This setup provides an excellent experience with this model. Use Ollama or vLLM for best results.
With 80 GB VRAM on NVIDIA H100 PCIe 80GB, use Q8_0 for most models — it is near-lossless and you have the memory for it. For 70B+ models, Q6_K offers excellent quality. Reserve Q4_K_M for 100B+ models or when you need maximum context length.
NVIDIA H100 PCIe 80GB already has strong memory bandwidth, so the next limit is often memory capacity and context headroom rather than raw decode speed. For local LLMs, fit first and bandwidth second is the right mental model.
NVIDIA H100 PCIe 80GB supports up to 4× GPU scaling via PCIe. With 4× GPUs, you get 320 GB effective memory with a 0.78× scaling factor per GPU. This enables running models like Qwen 3.5 397B A17B and Kimi K2.5 that don't fit on a single card.
NVIDIA H100 PCIe 80GB uses PCIe for multi-GPU communication, which has approximately 22% scaling overhead. For best multi-GPU performance, consider NVLink-equipped variants.
Usually yes. If you want to run 2-4× NVIDIA H100 PCIe 80GB for local AI, the bottleneck often becomes the platform, not the card. Workstation and server boards give you more CPU PCIe lanes, better x16 slot wiring, more spacing between cards, stronger power delivery, and usually more RAM capacity. Consumer x8/x8 layouts can work, but they are a common weak point in multi-GPU builds.
Compare with similar
