Outmake Models
Models

Spark X2.5 1.7B

XHToken · August 2026

Get Hugging Face

A 1.7B chat model from XHToken.

Downloads
247K
Likes
62
Parameters
1.7B
Context
1M
Licence
apache-2.0

Runs great on M4 · 16 GB

The Q8_0 size from Mungert is the one to get: a 1.8 GB download that uses 6.3 GB of memory, and writes about 29 words a second.

Spark X2.5 1.7B runs well from a Mac with an M1 and 8 GB of memory. 59 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From Mungert/Spark-X2.5-1.7B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
IQ3_XXS0.8 GB1.0 GBestimatedOpen
IQ3_XS0.8 GB1.0 GBestimatedOpen
IQ3_S0.8 GB1.0 GBestimatedOpen
IQ3_M0.9 GB1.1 GBestimatedOpen
Q3_K_S0.9 GB1.1 GBestimatedOpen
Q4_01.0 GB1.2 GBestimatedOpen
Q3_K_M1.0 GB1.2 GBestimatedOpen
IQ4_XS1.0 GB1.2 GBestimatedOpen
Q3_K_L1.0 GB1.2 GBestimatedOpen
IQ4_NL1.1 GB1.2 GBestimatedOpen
Q4_11.1 GB1.3 GBestimatedOpen
Q4_K_S1.1 GB1.3 GBestimatedOpen
Q4_0_L1.1 GB1.3 GBestimatedOpen
Q4_K_M1.1 GB1.3 GBestimatedOpen
Q5_01.2 GB1.4 GBestimatedOpen
Q4_K_L1.2 GB1.4 GBestimatedOpen
Q4_1_L1.2 GB1.4 GBestimatedOpen
Q5_K_S1.3 GB1.4 GBestimatedOpen
Q5_K_M1.3 GB1.5 GBestimatedOpen
Q5_0_L1.3 GB1.5 GBestimatedOpen
Q5_11.3 GB1.5 GBestimatedOpen
Q5_K_L1.3 GB1.5 GBestimatedOpen
Q5_1_L1.4 GB1.6 GBestimatedOpen
Q6_K_M1.4 GB1.6 GBestimatedOpen
Q6_K_L1.5 GB1.7 GBestimatedOpen
Q8_01.8 GB2.0 GBestimatedOpen
Q4_K2.3 GB2.5 GBestimatedOpen
Q6_K2.5 GB2.7 GBestimatedOpen
BF163.4 GB3.6 GBestimatedOpen
F163.4 GB3.6 GBestimatedOpen

Run It Anywhere

Spark X2.5 1.7B is an open model, so any app that runs GGUF files can run it. These take the Q8_0 size from Mungert straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/Mungert/Spark-X2.5-1.7B-GGUF:Q8_0
  • llama.cppMac, Windows, Linuxllama-server -hf Mungert/Spark-X2.5-1.7B-GGUF:Q8_0
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/Mungert/Spark-X2.5-1.7B-GGUF:Q8_0
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBF163.4 GB3.6 GB
8 GBF163.4 GB3.6 GB
10 GBF163.4 GB3.6 GB
11 GBF163.4 GB3.6 GB
12 GBF163.4 GB3.6 GB
16 GBF163.4 GB3.6 GB
20 GBF163.4 GB3.6 GB
24 GBF163.4 GB3.6 GB
32 GBF163.4 GB3.6 GB
48 GBF163.4 GB3.6 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 GBQ8_0Runs great6.3 GB46
M6 · 24 GBQ8_0Runs great11 GB46
M6 · 32 GBQ8_0Runs great11 GB46
M5 · 16 GBQ8_0Runs great6.3 GB43
M5 · 24 GBQ8_0Runs great11 GB43
M5 · 32 GBQ8_0Runs great11 GB43
M5 Pro · 24 GBQ8_0Runs great11 GB81
M5 Pro · 48 GBQ8_0Runs great11 GB81
M5 Pro · 64 GBQ8_0Runs great11 GB81
M5 Max · 36 GBQ8_0Runs great11 GB114
M5 Max · 48 GBQ8_0Runs great11 GB144
M5 Max · 64 GBQ8_0Runs great11 GB144
M5 Max · 128 GBQ8_0Runs great11 GB144
M5 Ultra · 96 GBQ8_0Runs great11 GB237
M5 Ultra · 256 GBQ8_0Runs great11 GB237
M5 Ultra · 512 GBQ8_0Runs great11 GB237
M4 · 16 GBQ8_0Runs great6.3 GB29
M4 · 24 GBQ8_0Runs great11 GB29
M4 · 32 GBQ8_0Runs great11 GB29
M4 Pro · 24 GBQ8_0Runs great11 GB55
M4 Pro · 48 GBQ8_0Runs great11 GB55
M4 Pro · 64 GBQ8_0Runs great11 GB55
M4 Max · 36 GBQ8_0Runs great11 GB71
M4 Max · 48 GBQ8_0Runs great11 GB83
M4 Max · 64 GBQ8_0Runs great11 GB83
M4 Max · 96 GBQ8_0Runs great11 GB71
M4 Max · 128 GBQ8_0Runs great11 GB83
M3 · 8 GBQ8_0Runs great3.0 GB26
M3 · 16 GBQ8_0Runs great6.3 GB26
M3 · 24 GBQ8_0Runs great11 GB26
M3 Pro · 18 GBQ8_0Runs great6.3 GB36
M3 Pro · 36 GBQ8_0Runs great11 GB36
M3 Max · 36 GBQ8_0Runs great11 GB62
M3 Max · 48 GBQ8_0Runs great11 GB75
M3 Max · 64 GBQ8_0Runs great11 GB75
M3 Max · 96 GBQ8_0Runs great11 GB62
M3 Max · 128 GBQ8_0Runs great11 GB75
M3 Ultra · 96 GBQ8_0Runs great11 GB109
M3 Ultra · 256 GBQ8_0Runs great11 GB109
M3 Ultra · 512 GBQ8_0Runs great11 GB109
M2 · 8 GBQ8_0Runs great3.0 GB26
M2 · 16 GBQ8_0Runs great6.3 GB26
M2 · 24 GBQ8_0Runs great11 GB26
M2 Pro · 16 GBQ8_0Runs great6.3 GB46
M2 Pro · 32 GBQ8_0Runs great11 GB46
M2 Max · 32 GBQ8_0Runs great11 GB74
M2 Max · 64 GBQ8_0Runs great11 GB74
M2 Max · 96 GBQ8_0Runs great11 GB74
M2 Ultra · 64 GBQ8_0Runs great11 GB108
M2 Ultra · 128 GBQ8_0Runs great11 GB108
M2 Ultra · 192 GBQ8_0Runs great11 GB108
M1 · 8 GBQ8_0Runs great3.0 GB17
M1 · 16 GBQ8_0Runs great6.3 GB17
M1 Pro · 16 GBQ8_0Runs great6.3 GB40
M1 Pro · 32 GBQ8_0Runs great11 GB40
M1 Max · 32 GBQ8_0Runs great11 GB60
M1 Max · 64 GBQ8_0Runs great11 GB60
M1 Ultra · 64 GBQ8_0Runs great11 GB80
M1 Ultra · 128 GBQ8_0Runs great11 GB80

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

Packaged by

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

Information

Maker
XHToken
Released
August 2026
Parameters
1.7B
Longest chat
1M tokens
Reads pictures
No
Licence
apache-2.0
Downloads this month
247,065
Likes
62
Sizes
30, 0.8 to 3.4 GB
Kind
Maker's release

Badge for the model card

Shows on Spark X2.5 1.7B'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