How much VRAM does BLOOM 176B need?
BLOOM 176B needs about 105.1 GB of VRAM at Q4_K_M, or 183.6 GB at Q8_0 and 345.0 GB at FP16. No consumer card holds it at Q4_K_M. The smallest fit in our set is the H200 141GB (141 GB). Weights plus runtime overhead; with only 2K of context the cache stays small.
Weights-plus-runtime footprint across the six most common quantizations. KV cache is context-dependent and comes on top. The 2K window is short, so weights dominate the budget; size it precisely in the GPU Memory Calculator.
Can your GPU run BLOOM 176B?
Pick your card — the answer below is computed by the same engine that produces every figure on this page.
H200 141GB runs BLOOM 176B at Q4_K_M with 28.9 GB to spare; comfortable at this size, where most cards cannot hold the weights at all.
Weights plus runtime overhead. KV cache is not included — it depends on context length and this model’s published architecture.
| Weights (Q4_K_M) | 99.4 GB |
| Runtime overhead | 5.7 GB |
| Estimated total | 105.1 GB |
| Usable VRAM (95% of 141 GB) | 133.9 GB |
Across every GPU we track
- RTX 3060 12GBNot enough-93.7 GB
- RTX 4060 Ti 16GBNot enough-89.9 GB
- RTX 4070 Ti SUPERNot enough-89.9 GB
- RTX 3090Not enough-82.3 GB
- RTX 4090Not enough-82.3 GB
- RTX 5090Not enough-74.7 GB
- A100 40GBNot enough-67.1 GB
- RTX A6000Not enough-59.5 GB
- L40SNot enough-59.5 GB
- A100 80GBNot enough-29.1 GB
- H100 80GBNot enough-29.1 GB
- H200 141GBRecommended+28.9 GB
Estimates, not guarantees: real usage moves with runtime, driver, batch size and context. Size a specific context window in the GPU Memory Calculator.
How was this number calculated? Every figure here comes from Bitpute’s documented calculation methodology — parameters, precision, quantization, runtime overhead and usable VRAM, each formula written out in full.
Memory by quantization
| Quant | Weights | + runtime overhead |
|---|---|---|
| FP16 | 327.8 GB | 345.0 GB |
| Q8_0 | 174.2 GB | 183.6 GB |
| Q6_K | 134.4 GB | 141.9 GB |
| Q5_K_M | 116.6 GB | 123.2 GB |
| Q4_K_M | 99.4 GB | 105.1 GB |
| Q4_0 | 92.2 GB | 97.6 GB |
Overhead = 0.75 GB + 5% of weights (CUDA context, buffers). Add KV cache on top. With only 2K of context, weights dominate the budget.
Which quantization should you actually run?
BLOOM 176B is too large for a 12–16 GB card. On H200 141GB, Q4_K_M (105.1 GB) fits with plenty of headroom for long context — the realistic single-GPU entry point. Higher quality (Q6_K 141.9 GB, FP16 345.0 GB) needs a 48 GB+ card or multiple GPUs — for occasional use, renting is usually cheaper than buying.
Single-GPU fit at Q4_K_M (105.1 GB + KV)
RTX 3060 12GBRTX 4060 Ti 16GBRTX 4070 Ti SUPERRTX 3090RTX 4090RTX 5090A100 40GBRTX A6000L40SA100 80GBH100 80GBH200 141GB
Where the memory goes
Weights and overhead are exact for Q4_K_M. The KV bar is a generic 32-layer transformer at 8K; at this scale the weights dominate, but long context still adds gigabytes — size it in the calculator is for.
Common questions
How much VRAM does BLOOM 176B need?
BLOOM 176B needs about 105.1 GB of VRAM at Q4_K_M, 183.6 GB at Q8_0, or 345.0 GB at FP16. That is weights plus about 0.75 GB of runtime overhead and 5% of weight size; KV cache is additional and depends on context length.
What GPU can run BLOOM 176B?
At Q4_K_M the smallest card in our set that fits is the H200 141GB (141 GB usable at 95%). At FP16 no single GPU in our set is sufficient.
Can BLOOM 176B run on 24 GB?
Not at Q4_K_M: it needs about 105.1 GB, more than the ~22.8 GB usable on a 24 GB card. Use a larger card, multiple GPUs, or offload.
Same memory footprint as 1 other checkpoint
BLOOM 176B has an identical parameter count to this, so every figure on this page applies to it unchanged. They are separate checkpoints of the same architecture at the same size, so the memory budget is identical — check each model card for capability differences.
More in the BLOOM family
BLOOM 560MBLOOM 560M zBLOOM 1.7BBLOOM 1.7B zBLOOM 3BBLOOM 3B z