B200
30 records across 12 models (quants: DEP4, FP4, FP8, Q4_0, unspecified).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B200 | AFM-4.5B-Base | unspecified | vLLM | 13820 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | AFM-4.5B-Base | unspecified | vLLM | 13657 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | DeepSeek-R1-0528 DSV1 MoE (0528) | FP8 | vLLM | 2421 | – | – | – | – | 2025-11 | community | – | source |
| B200 | DeepSeek-R1-0528 DSV1 MoE (0528) | FP8 | vLLM | 1994 | – | – | – | – | 2025-11 | community | – | source |
| B200 | DeepSeek-R1-0528-FP4-v2 DSV1 MoE (0528) | FP4 | vLLM | 2895 | – | – | – | – | 2025-11 | community | – | source |
| B200 | DeepSeek-R1-0528-FP4-v2 DSV1 MoE (0528) | FP4 | vLLM | 2375 | – | – | – | – | 2025-11 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 245996 | – | – | – | 32768 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 245556 | – | – | – | 8192 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 236952 | – | – | – | 8192 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 235791 | – | – | – | 32768 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 204764 | – | – | – | 131072 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 194012 | – | – | – | 131072 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 1303 | – | – | – | 32768 | 2026-10-03 | community | – | source |
| B200 | DSV4.1-Flash | DEP4 | vLLM | 1281 | – | – | – | 32768 | 2026-10-03 | community | – | source |
| B200 | facebook/cwm | unspecified | vLLM | 1766 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | facebook/cwm | unspecified | vLLM | 1761 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | GLM-4-9B-Chat | unspecified | vLLM | 7591 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | GLM-4-9B-Chat | unspecified | vLLM | 7481 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | Jais-2-8B-Chat | unspecified | vLLM | 5359 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | Jais-2-8B-Chat | unspecified | vLLM | 5324 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | llama 7B | Q4_0 | llama.cpp (CUDA) | 315.85 | 11510.56 | – | – | tg128 | 2025-10-25 | community | – | source |
| B200 | llama 7B | Q4_0 | llama.cpp (CUDA) | 297.68 | 10531.52 | – | – | tg128 | 2025-10-25 | community | – | source |
| B200 | Mellum2-12B-A2.5B-Base | unspecified | vLLM | 16604 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | Mellum2-12B-A2.5B-Base | unspecified | vLLM | 16498 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | phi-2 | unspecified | vLLM | 5155 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | phi-2 | unspecified | vLLM | 5144 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | pythia-12b | unspecified | vLLM | 2912 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | pythia-12b | unspecified | vLLM | 2911 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | Seed-OSS-36B-Base | unspecified | vLLM | 1542 | – | – | – | 8192 | 2026-10-01 | community | – | source |
| B200 | Seed-OSS-36B-Base | unspecified | vLLM | 1540 | – | – | – | 8192 | 2026-10-01 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085