WHAT THIS CARD IS WORTH
B200 holds 390 of the models in our catalogue and is, in practice, a Q4_K_S card — the largest it takes is DBRX 132B at Q4_K_S.
TOKENS/SEC PER $100
0.1
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
0.30 GB
what memory costs on this card
THE BIGGEST IT TAKES
AGAINST CARDS OF SIMILAR MEMORY
B20090 GB$30,0000.1 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 B200 — 90 GB VRAM.
▸ B200 SPEC
- BRAND
- NVIDIA
- VRAM
- 90 GB HBM3e
- BANDWIDTH
- 4100 GB/s
- FP16 COMPUTE
- 1191.2 TFLOPS
- FP32 COMPUTE
- 74.5 TFLOPS
- CUDA CORES
- 18,944
- TENSOR CORES
- 592
- TDP
- 1000 W
- ARCHITECTURE
- Blackwell
- MSRP
- $30000
▸ AI CAPABILITY
390/ 449 models @ Q4
With 90 GB VRAM and 4100 GB/s bandwidth, this GPU handles models up to 124B parameters.
Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~469 tok/s.
RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR B200
390 FIT · SHOWING 20| MODEL | SIZE | VRAM Q4 | TOK/S | AVG |
|---|---|---|---|---|
| Pixtral Large 124B | 124B | 76.3 GB | 26 | 39.3 |
| Nemotron 3 Super 120B-A12B | 123.61B | 76.0 GB | 273 | 53.2 |
| Mistral-Large 123B | 123B | 75.7 GB | 27 | 33.5 |
| Devstral 2 123B | 123B | 75.7 GB | 27 | 38.1 |
| Qwen 3.5 122B A10B | 122B | 75.1 GB | 328 | 56.8 |
| Nemotron 3 Super 120B | 120B | 73.8 GB | 273 | 57.3 |
| Mistral Small 4 119B | 119B | 73.2 GB | 505 | 50.2 |
| GPT-OSS 120B | 117B | 72.0 GB | 643 | 54.1 |
| Command A 111B | 111B | 68.3 GB | 30 | 27.6 |
| GLM 4.5 Air | 110B | 67.7 GB | 273 | 51.0 |
| Qwen 1.5 110B | 110B | 67.7 GB | 30 | 33.4 |
| Llama 4 Scout 17B-16E | 109B | 67.1 GB | 193 | 33.9 |
| Cogito v2 109B MoE | 109B | 67.1 GB | 193 | — |
| Ling 2.6 Flash | 107.49B | 66.2 GB | 443 | 36.8 |
| Sarvam 105B | 105B | 64.7 GB | 31 | 48.0 |
| Command-R+ 104B | 104B | 64.1 GB | 32 | 52.7 |
| Llama-3.2-90B-Vision-Instruct | 90B | 55.5 GB | 36 | 48.5 |
| Hunyuan A13B | 80B | 49.4 GB | 252 | 81.1 |
| Qwen3-Coder-Next | 80B | 49.4 GB | 1093 | 43.0 |
| Qwen3-Next 80B A3B | 80B | 49.4 GB | 1093 | 49.0 |