Outmake Models
Models

Laguna XS 2.1

poolside · June 2026

Get Hugging Face

A 33B, 3B active chat model from poolside.

Rating
3.1
Downloads
1.2M
Likes
158
Parameters
33B, 3B active
Context
256K
Licence
openmdw-1.1

Runs well on M4 · 16 GB

The IQ2_XXS size from bartowski is the one to get: a 9.4 GB download that uses 10 GB of memory, and writes about 7.4 words a second.

Ratings

3.1 out of 5, across the tests below.

Software fixesSWE-bench Verified
70.9%
Terminal tasksTerminal-Bench 2.0
37.5%

SWE-bench Verified: real GitHub issues fixed so the project's tests pass. Terminal-Bench 2.0: jobs finished alone in a terminal. Scores are the ones filed on the model's Hugging Face page or the board's own table, read 2026-11-05; some come from the maker.

Laguna XS 2.1 runs well from a Mac with an M1 and 16 GB of memory. 56 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From poolside/Laguna-XS-2.1-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q4_K_M20 GB21 GBestimatedOpen
BF1667 GB68 GBestimatedOpen

Run It Anywhere

Laguna XS 2.1 is an open model, so any app that runs GGUF files can run it. These take the IQ2_XXS size from bartowski straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/Laguna-XS-2.1-GGUF:IQ2_XXS
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/Laguna-XS-2.1-GGUF:IQ2_XXS
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/Laguna-XS-2.1-GGUF:IQ2_XXS
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 21 GB
8 GBNone worth runningneeds 21 GB
10 GBNone worth runningneeds 21 GB
11 GBNone worth runningneeds 21 GB
12 GBNone worth runningneeds 21 GB
16 GBNone worth runningneeds 21 GB
20 GBNone worth runningneeds 21 GB
24 GBQ4_K_M20 GB21 GB
32 GBQ4_K_M20 GB21 GB
48 GBQ4_K_M20 GB21 GB

The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.

On every Mac
MacSizeFitMemoryWrites
M6 · 16 GBIQ2_XXSRuns well10 GB11
M6 · 24 GBQ3_K_SRuns well16 GB10
M6 · 32 GBQ4_K_MRuns well21 GB9.4
M5 · 16 GBIQ2_XXSRuns well10 GB10
M5 · 24 GBQ3_K_SRuns well16 GB9.3
M5 · 32 GBQ4_K_MRuns well21 GB8.6
M5 Pro · 24 GBQ3_K_SRuns well16 GB18
M5 Pro · 48 GBQ4_K_MRuns great22 GB17
M5 Pro · 64 GBQ6_K_LRuns great31 GB15
M5 Max · 36 GBIQ3_MRuns great17 GB26
M5 Max · 48 GBQ4_K_MRuns great22 GB32
M5 Max · 64 GBQ6_K_LRuns great31 GB29
M5 Max · 128 GBBF16Runs great69 GB24
M5 Ultra · 96 GBQ8_0Runs great42 GB57
M5 Ultra · 256 GBBF16Runs great92 GB45
M5 Ultra · 512 GBBF16Runs great92 GB45
M4 · 16 GBIQ2_XXSRuns well10 GB7.4
M4 · 24 GBQ3_K_SRuns well16 GB6.9
M4 · 32 GBQ4_K_MRuns well21 GB6.4
M4 Pro · 24 GBQ3_K_SRuns well16 GB14
M4 Pro · 48 GBQ4_K_MRuns great22 GB13
M4 Pro · 64 GBQ6_K_LRuns great31 GB12
M4 Max · 36 GBIQ3_MRuns great17 GB19
M4 Max · 48 GBQ4_K_MRuns great22 GB22
M4 Max · 64 GBQ6_K_LRuns great31 GB21
M4 Max · 96 GBQ8_0Runs great42 GB18
M4 Max · 128 GBBF16Runs great69 GB18
M3 · 8 GBIQ1_SToo big32 GB
M3 · 16 GBIQ2_XXSRuns well10 GB6.3
M3 · 24 GBQ3_K_SRuns well16 GB5.9
M3 Pro · 18 GBIQ1_MRuns great8.5 GB9.4
M3 Pro · 36 GBIQ3_MRuns great17 GB8.4
M3 Max · 36 GBIQ3_MRuns great17 GB15
M3 Max · 48 GBQ4_K_MRuns great22 GB19
M3 Max · 64 GBQ6_K_LRuns great31 GB17
M3 Max · 96 GBQ8_0Runs great42 GB14
M3 Max · 128 GBBF16Runs great69 GB14
M3 Ultra · 96 GBQ8_0Runs great42 GB31
M3 Ultra · 256 GBBF16Runs great92 GB25
M3 Ultra · 512 GBBF16Runs great92 GB25
M2 · 8 GBIQ1_SToo big32 GB
M2 · 16 GBIQ2_XXSRuns well10 GB6.3
M2 · 24 GBQ3_K_SRuns well16 GB5.9
M2 Pro · 16 GBIQ2_XXSRuns well10 GB12
M2 Pro · 32 GBQ4_K_MRuns well21 GB10
M2 Max · 32 GBQ4_K_MRuns well21 GB18
M2 Max · 64 GBQ6_K_LRuns great31 GB17
M2 Max · 96 GBQ8_0Runs great42 GB18
M2 Ultra · 64 GBQ6_K_LRuns great31 GB28
M2 Ultra · 128 GBBF16Runs great69 GB25
M2 Ultra · 192 GBBF16Runs great92 GB25
M1 · 8 GBIQ1_SToo big32 GB
M1 · 16 GBIQ2_XXSRuns well10 GB4.3
M1 Pro · 16 GBIQ2_XXSRuns well10 GB11
M1 Pro · 32 GBQ4_K_MRuns well21 GB9.6
M1 Max · 32 GBQ4_K_MRuns well21 GB16
M1 Max · 64 GBQ6_K_LRuns great31 GB15
M1 Ultra · 64 GBQ6_K_LRuns great31 GB23
M1 Ultra · 128 GBBF16Runs great69 GB21

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Packaged by

5 repositories on Hugging Face carry this model for llama.cpp. Outmake reads them all as one model.

Information

Maker
poolside
Released
June 2026
Parameters
33B, 3B active
Longest chat
256K tokens
Reads pictures
No
Licence
openmdw-1.1
Downloads this month
1,222,131
Likes
158
Sizes
2, 20 to 67 GB
Kind
Maker's release

Badge for the model card

Shows on Laguna XS 2.1's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta