# Laguna S 2.1 on an A100 40GB (SXM)

No. Even Q3_K_M needs 50.3 GiB against 36.8 GiB usable.

Canonical page: https://makerportal.ai/lab/llm-vram/laguna-s-2-1/a100-40gb
Page title: Laguna S 2.1 VRAM on A100 40GB — does not fit

This markdown document and the HTML page above are rendered from the same solved values at build time, by the same functions. Nothing here is written by a language model and nothing is fetched at request time.

## Key figures

- **Verdict:** Does not fit — 36.8 GiB usable of 40 GB
- **Usable memory:** 36.8 GiB — 92% of the card's 40 GB — an assumption for a dedicated card, not a single fraction applied to every device
- **Weights at FP16:** 219.0 GiB — 117.6B parameters
- **Weights (Q4_K_M):** 66.8 GiB — 219.0 GiB at FP16 · 117.6B params · 6 of 6 quant rows are measured files
- **KV cache:** 192.0 KiB per token at FP16 — 48 layers × 8 KV heads × 128 dimensions
- **Speed ceiling:** 59.5 tok/s — offloaded — nothing fits in device memory
- **Largest context that fits:** none — the weights do not fit
- **Trained context:** 1,048,576 — the cap the KV-cache figures are held to
- **Card bandwidth:** 1555 GB/s — 40 GB HBM2

## Every quant, against this card

Every quantization of Laguna S 2.1 against the 36.8 GiB usable on an A100 40GB (SXM). "Measured" rows are the byte size of a real published file; "nominal" rows are the parameter count times the published bits-per-weight.

| Precision | Bits/weight | Weights | Source | Fits | Max context | Ceiling |
|---|---|---|---|---|---|---|
| FP16 / BF16 | 16.00 | 219.0 GiB | measured file | short by 182.2 GiB | — | 7.1 tok/s offloaded |
| Q8_0 | 8.51 | 116.4 GiB | measured file | short by 79.6 GiB | — | 15.3 tok/s offloaded |
| Q6_K | 6.92 | 94.7 GiB | measured file | short by 57.9 GiB | — | 20.3 tok/s offloaded |
| Q5_K_M | 5.72 | 78.2 GiB | measured file | short by 41.4 GiB | — | 26.8 tok/s offloaded |
| Q4_K_M | 4.88 | 66.8 GiB | measured file | short by 30.0 GiB | — | 34.6 tok/s offloaded |
| Q3_K_M | 3.68 | 50.3 GiB | measured file | short by 13.5 GiB | — | 59.5 tok/s offloaded |

## Questions this page answers

### How much VRAM does Laguna S 2.1 need?

219.0 GiB for the weights at FP16 — 117.6B parameters at two bytes each — and 66.8 GiB at Q4_K_M. The KV cache is on top of that and is not a fixed number: this model spends 192.0 KiB per token of context on its full-attention layers, so 8,192 tokens costs a further 456 MiB. An A100 40GB (SXM) makes 36.8 GiB of its 40 GB available on the assumption below.

### Can an A100 40GB (SXM) run Laguna S 2.1?

No. Even Q3_K_M needs 50.3 GiB against 36.8 GiB usable. Holding the smallest quant here would need a card with about 55 GB. Splitting the model across the PCIe bus is possible and slow — see the offload figure on this page.

### How fast will Laguna S 2.1 run on an A100 40GB (SXM)?

It cannot run in this card's memory alone, so the speed is set by whatever bus the offloaded part is read across, not by the 1555 GB/s of the card. At an assumed 90 GB/s of host memory bandwidth the ceiling is 59.5 tok/s at Q3_K_M, against 409.2 tok/s if it were resident — 1555 GB/s over the 3.80 GB one token reads at Q3_K_M, priced at the 8k reference context. That is the weights the model routes through plus one pass over the cache, against a 54.04 GB weight file.

### Why does the context length change how much memory Laguna S 2.1 needs?

Because the KV cache holds one key and one value vector per token, per layer, for the whole conversation, and it is allocated separately from the weights. This model has 48 layers and 8 key/value heads of 128 dimensions, shared across 48 query heads — grouped-query attention, which divides the cache by 6. That works out at 192.0 KiB per token, except that 36 of the 48 layers use a 512-token sliding window and stop growing there. Parameter count tells you nothing about this number.

## Method and limits

Weight bytes are the byte size of the real published file wherever one exists, and the model's exact parameter count times the published llama.cpp bits-per-weight where it does not. That distinction is on every row above and it matters at both ends: a sub-1B model's Q4_K_M file runs a third larger than the nominal figure because k-quants keep its embedding tables at higher precision, and an already-4-bit release cannot be quantized upward at all. The KV cache is 2 · layers · kv_heads · head_dim · bytes per token, summed over layers with each sliding-window layer capped at its window. The speed figure is a roofline bound, not a benchmark: bandwidth divided by bytes read per token, which no runtime exceeds and every runtime falls short of. The usable fraction of card memory is an assumption: 92% here, which is what this lane assumes for a dedicated card — the other class assumes 75%, so it is not one number applied to every device. Because nothing here fits, a second assumed input is in play — the 90 GB/s of host memory bandwidth the offload ceiling is priced at, stated above. Those two are the assumed inputs; every other figure is computed from the model config and the card's published specification. Nothing on this page is written by a language model.

## Related tool

[LLM VRAM & KV-Cache Footprint Calculator](https://makerportal.ai/lab/llm-vram-kvcache-calculator) — Interactive VRAM planner — change the model geometry, quantization, context length and card capacity and read the weight bytes, KV-cache bytes and the bandwidth ceiling on decode speed.

---

Source: MakerPortal — https://makerportal.ai/lab/llm-vram/laguna-s-2-1/a100-40gb. Free to quote and cite with attribution and a link to the canonical page.