# Laguna XS 2.1 on a GeForce RTX 5080

Only technically. Q3_K_M weights fit in 14.7 GiB, but they leave room for just 4,068 tokens of context — below anything useful.

Canonical page: https://makerportal.ai/lab/llm-vram/laguna-xs-2-1/rtx-5080
Page title: Laguna XS 2.1 VRAM on RTX 5080 — barely fits at Q3_K_M

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:** Too tight — 14.7 GiB usable of 16 GB
- **Usable memory:** 14.7 GiB — 92% of the card's 16 GB — an assumption for a dedicated card, not a single fraction applied to every device
- **Weights at FP16:** 62.3 GiB — 33.4B parameters
- **Weights (Q4_K_M):** 19.1 GiB — 62.3 GiB at FP16 · 33.4B params · 6 of 6 quant rows are measured files
- **KV cache:** 160.0 KiB per token at FP16 — 40 layers × 8 KV heads × 128 dimensions
- **Speed ceiling:** 670 tok/s — 960 GB/s ÷ bytes read per token
- **Largest context that fits:** 4,068 tokens at Q3_K_M
- **Trained context:** 262,144 — the cap the KV-cache figures are held to
- **Card bandwidth:** 960 GB/s — 16 GB GDDR7

## Every quant, against this card

Every quantization of Laguna XS 2.1 against the 14.7 GiB usable on a GeForce RTX 5080. "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 | 62.3 GiB | measured file | short by 47.6 GiB | — | 20.2 tok/s offloaded |
| Q8_0 | 8.52 | 33.2 GiB | measured file | short by 18.4 GiB | — | 44.1 tok/s offloaded |
| Q6_K | 6.95 | 27.0 GiB | measured file | short by 12.3 GiB | — | 58.6 tok/s offloaded |
| Q5_K_M | 5.74 | 22.4 GiB | measured file | short by 7.6 GiB | — | 78.5 tok/s offloaded |
| Q4_K_M | 4.92 | 19.1 GiB | measured file | short by 4.4 GiB | — | 102.3 tok/s offloaded |
| Q3_K_M | 3.73 | 14.5 GiB | measured file | yes | 4,068 | 670 tok/s |

## Questions this page answers

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

62.3 GiB for the weights at FP16 — 33.4B parameters at two bytes each — and 19.1 GiB at Q4_K_M. The KV cache is on top of that and is not a fixed number: this model spends 160.0 KiB per token of context on its full-attention layers, so 8,192 tokens costs a further 380 MiB. A GeForce RTX 5080 makes 14.7 GiB of its 16 GB available on the assumption below.

### Can a GeForce RTX 5080 run Laguna XS 2.1?

Only technically. Q3_K_M weights fit in 14.7 GiB, but they leave room for just 4,068 tokens of context — below anything useful. That is the weights and the KV cache together, against 14.7 GiB of usable memory.

### How fast will Laguna XS 2.1 run on a GeForce RTX 5080?

No faster than 670 tokens/second at Q3_K_M, and in practice below it. Decoding is memory-bound: every token reads the 8% of weights this MoE routes to 1.1 GiB of weights plus the cache, and this card moves 960 GB/s. That division is the ceiling — no kernel, runtime or driver beats it, and a real runtime typically reaches 60–80% of it. That figure assumes the weights are resident, which here leaves almost no room for context — the speed is not the binding problem.

### Why does the context length change how much memory Laguna XS 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 40 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 160.0 KiB per token, except that 30 of the 40 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. That is the only assumed input on this page; 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-xs-2-1/rtx-5080. Free to quote and cite with attribution and a link to the canonical page.