Yes — with 17.1 GB to spare

LLM-jp 4 33B Thinking at Q4_K_M fits your A100 40 GB entirely on the GPU at 8K context, at an estimated 50 tokens per second. There is room for its full 64K window.

Fully on GPU 8K context Q4_K_M · 18.7 GB Apache 2.0 Released 14 Aug 2026 New this week Not in the Ollama library

Japan’s national-institute reasoning model, Japanese and English. A plain dense Llama-style 33B: Q4 is a tight 24 GB fit.

What hardware do I need for LLM-jp 4 33B Thinking? →

The VRAM budget

weights 18.7 GB

Weights 18.7 GB KV cache @ 8K 2.00 GB Runtime overhead 0.6 GB Free 17.1 GB of 38.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	32.9 GB	35.5 GB	19K	29	−0.1% ppl	Long context
Q6_K	25.4 GB	28.0 GB	49K	37	−0.4% ppl	Long context
Q5_K_M	21.9 GB	24.5 GB	63K	43	−0.8% ppl	Long context
Q4_K_M	18.7 GB	21.3 GB	64K	50	−1.9% ppl	Recommended
Q3_K_M	15.1 GB	17.7 GB	64K	62	−5.4% ppl	Long context
Q2_K	12.9 GB	15.5 GB	64K	73	−15% ppl	Long context

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.

How to run it

llama.cpp LM Studio vLLM

terminal

$ llama-server \
    -hf llm-jp/llm-jp-4-33b-thinking:Q4_K_M \
    -c 8192 -ngl 99

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 18.7 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.

02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.

03There is room to go to the model's full 64K context on this card.

See all models for this rig Compare with another model