RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can internlm2 5 7b chat run on the H200?

Yes. At 8K context the best fit is F16, using about 16.0 GB of the 134 GB usable on the H200. Estimated speed ceiling: ~310 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •3.9 GB Fits4.4 GB Fits7.4 GB Fits1597
3-bit (Q3_K_M)4.6 GB Fits5.1 GB Fits8.1 GB Fits1269
4-bit (Q4_K_M)5.5 GB Fits6.0 GB Fits9.0 GB Fits1015
5-bit (Q5_K_M)6.2 GB Fits6.7 GB Fits9.7 GB Fits871
6-bit (Q6_K) •7.0 GB Fits7.5 GB Fits10.5 GB Fits756
8-bit (Q8_0) •8.7 GB Fits9.2 GB Fits12.2 GB Fits584
16-bit (F16/BF16) •15.5 GB Fits16.0 GB Fits19.0 GB Fits310

• real GGUF file size from legraphista/internlm2_5-7b-chat-IMat-GGUF. Speed is a bandwidth ceiling (4800 GB/s), not a benchmark. Methodology.

More

internlm2 5 7b chat on every GPU and Mac · Every model that runs on the H200 · Other models that fit 64 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (internlm/internlm2_5-7b-chat), NVIDIA specs. Data as of 2026-11-17.