WHAT THIS CARD IS WORTH
H100 CNX holds 382 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is Nemotron 3 Super 120B-A12B at IQ4_XS.
TOKENS/SEC PER $100
0.2
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
0.32 GB
what memory costs on this card
THE BIGGEST IT TAKES
AGAINST CARDS OF SIMILAR MEMORY
H100 CNX80 GB$25,0000.2 tok/s per $100
Neighbours in memory rather than in price: memory decides whether a card can do the job at all, so two cards of the same size at different prices is the comparison you are making. Ordered by memory, then bandwidth — not by the value column, which is worked out from bandwidth and price alone and therefore rewards a cheap card whatever its software stack does to that bandwidth in practice. Read it as one input, not as a ranking.
▸ DEVICE UNDER TEST
NVIDIA H100 CNX — 80 GB VRAM.
▸ H100 CNX SPEC
- BRAND
- NVIDIA
- VRAM
- 80 GB HBM2e
- BANDWIDTH
- 2040 GB/s
- FP16 COMPUTE
- 215.4 TFLOPS
- FP32 COMPUTE
- 53.8 TFLOPS
- CUDA CORES
- 14,592
- TENSOR CORES
- 456
- TDP
- 350 W
- ARCHITECTURE
- Hopper
- MSRP
- $25000
▸ AI CAPABILITY
382/ 449 models @ Q4
With 80 GB VRAM and 2040 GB/s bandwidth, this GPU handles models up to 111B parameters.
Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~233 tok/s.
RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR H100 CNX
382 FIT · SHOWING 20| MODEL | SIZE | VRAM Q4 | TOK/S | AVG |
|---|---|---|---|---|
| Command A 111B | 111B | 68.3 GB | 15 | 27.6 |
| GLM 4.5 Air | 110B | 67.7 GB | 136 | 51.0 |
| Qwen 1.5 110B | 110B | 67.7 GB | 15 | 33.4 |
| Llama 4 Scout 17B-16E | 109B | 67.1 GB | 96 | 33.9 |
| Cogito v2 109B MoE | 109B | 67.1 GB | 96 | — |
| Ling 2.6 Flash | 107.49B | 66.2 GB | 221 | 36.8 |
| Sarvam 105B | 105B | 64.7 GB | 16 | 48.0 |
| Command-R+ 104B | 104B | 64.1 GB | 16 | 52.7 |
| Llama-3.2-90B-Vision-Instruct | 90B | 55.5 GB | 18 | 48.5 |
| Hunyuan A13B | 80B | 49.4 GB | 126 | 81.1 |
| Qwen3-Coder-Next | 80B | 49.4 GB | 544 | 43.0 |
| Qwen3-Next 80B A3B | 80B | 49.4 GB | 544 | 49.0 |
| NVLM-D 72B | 79.38B | 49.0 GB | 21 | 48.7 |
| InternVL3 78B | 78B | 48.2 GB | 21 | 80.6 |
| Qwen2.5-72B | 72.7B | 44.9 GB | 22 | 39.7 |
| Qwen2-VL 72B | 72.7B | 44.9 GB | 22 | 55.5 |
| Qwen 1.5 72B | 72B | 44.5 GB | 23 | 49.7 |
| Qwen2 Math 72B | 72B | 44.5 GB | 23 | 49.7 |
| Molmo 72B | 72B | 44.5 GB | 23 | 54.1 |
| DeepSeek R1 Distill Llama 70B | 70.6B | 43.6 GB | 23 | 42.4 |