Outmake Models
Models

Llama 3.1 Tulu 3 8B

Ai2 · November 2024

Get Hugging Face

A 8B chat model from Ai2.

Rating
2.4
Downloads
2.9K
Likes
4
Parameters
8B
Context
128K
Licence
llama3.1

Runs great on M4 · 16 GB

The Q6_K size from lmstudio-community is the one to get: a 6.6 GB download that uses 7.3 GB of memory, and writes about 6.3 words a second.

Ratings

2.4 out of 5, across people's blind votes on its answers.

Text
1221

LM Arena ratings of October 28, 2026 (CC BY 4.0): people vote between two answers without knowing which model wrote which. The bar ends at 1488, the best any open model has. The Arena tests the full model; a small size on a Mac answers a little worse.

Llama 3.1 Tulu 3 8B 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 lmstudio-community/Llama-3.1-Tulu-3-8B-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
Q3_K_L4.3 GB5.1 GBestimatedOpen
Q4_K_M4.9 GB5.6 GBestimatedOpen
Q6_K6.6 GB7.3 GBestimatedOpen
Q8_08.5 GB9.3 GBestimatedOpen

Run It Anywhere

Llama 3.1 Tulu 3 8B is an open model, so any app that runs GGUF files can run it. These take the Q6_K size from lmstudio-community straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/lmstudio-community/Llama-3.1-Tulu-3-8B-GGUF:Q6_K
  • llama.cppMac, Windows, Linuxllama-server -hf lmstudio-community/Llama-3.1-Tulu-3-8B-GGUF:Q6_K
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/lmstudio-community/Llama-3.1-Tulu-3-8B-GGUF:Q6_K
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBQ4_K_M4.9 GB5.6 GB
8 GBQ6_K6.6 GB7.3 GB
10 GBQ8_08.5 GB9.3 GB
11 GBQ8_08.5 GB9.3 GB
12 GBQ8_08.5 GB9.3 GB
16 GBQ8_08.5 GB9.3 GB
20 GBQ8_08.5 GB9.3 GB
24 GBQ8_08.5 GB9.3 GB
32 GBQ8_08.5 GB9.3 GB
48 GBQ8_08.5 GB9.3 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 GBQ6_KRuns great7.3 GB9.2
M6 · 24 GBQ8_0Runs great11 GB8.9
M6 · 32 GBQ8_0Runs great14 GB8.9
M5 · 16 GBQ6_KRuns great7.3 GB8.3
M5 · 24 GBQ8_0Runs great11 GB8.1
M5 · 32 GBQ8_0Runs great14 GB8.1
M5 Pro · 24 GBQ8_0Runs great11 GB16
M5 Pro · 48 GBQ8_0Runs great19 GB16
M5 Pro · 64 GBQ8_0Runs great19 GB16
M5 Max · 36 GBQ8_0Runs great14 GB24
M5 Max · 48 GBQ8_0Runs great19 GB31
M5 Max · 64 GBQ8_0Runs great19 GB31
M5 Max · 128 GBQ8_0Runs great19 GB31
M5 Ultra · 96 GBQ8_0Runs great19 GB59
M5 Ultra · 256 GBQ8_0Runs great19 GB59
M5 Ultra · 512 GBQ8_0Runs great19 GB59
M4 · 16 GBQ6_KRuns great7.3 GB6.3
M4 · 24 GBQ8_0Runs great11 GB6.1
M4 · 32 GBQ8_0Runs great14 GB6.1
M4 Pro · 24 GBQ8_0Runs great11 GB13
M4 Pro · 48 GBQ8_0Runs great19 GB13
M4 Pro · 64 GBQ8_0Runs great19 GB13
M4 Max · 36 GBQ8_0Runs great14 GB19
M4 Max · 48 GBQ8_0Runs great19 GB24
M4 Max · 64 GBQ8_0Runs great19 GB24
M4 Max · 96 GBQ8_0Runs great19 GB19
M4 Max · 128 GBQ8_0Runs great19 GB24
M3 · 8 GBQ3_K_LRuns well5.1 GB6.7
M3 · 16 GBQ6_KRuns great7.3 GB5.3
M3 · 24 GBQ8_0Runs great11 GB5.2
M3 Pro · 18 GBQ6_KRuns great7.9 GB7.9
M3 Pro · 36 GBQ8_0Runs great14 GB7.6
M3 Max · 36 GBQ8_0Runs great14 GB15
M3 Max · 48 GBQ8_0Runs great19 GB19
M3 Max · 64 GBQ8_0Runs great19 GB19
M3 Max · 96 GBQ8_0Runs great19 GB15
M3 Max · 128 GBQ8_0Runs great19 GB19
M3 Ultra · 96 GBQ8_0Runs great19 GB34
M3 Ultra · 256 GBQ8_0Runs great19 GB34
M3 Ultra · 512 GBQ8_0Runs great19 GB34
M2 · 8 GBQ3_K_LRuns well5.1 GB6.7
M2 · 16 GBQ6_KRuns great7.3 GB5.3
M2 · 24 GBQ8_0Runs great11 GB5.2
M2 Pro · 16 GBQ6_KRuns great7.3 GB10
M2 Pro · 32 GBQ8_0Runs great14 GB10
M2 Max · 32 GBQ8_0Runs great14 GB19
M2 Max · 64 GBQ8_0Runs great19 GB19
M2 Max · 96 GBQ8_0Runs great19 GB19
M2 Ultra · 64 GBQ8_0Runs great19 GB34
M2 Ultra · 128 GBQ8_0Runs great19 GB34
M2 Ultra · 192 GBQ8_0Runs great19 GB34
M1 · 8 GBQ3_K_LRuns well5.1 GB4.5
M1 · 16 GBQ6_KRuns great7.3 GB3.6
M1 Pro · 16 GBQ6_KRuns great7.3 GB9.9
M1 Pro · 32 GBQ8_0Runs great14 GB9.6
M1 Max · 32 GBQ8_0Runs great14 GB18
M1 Max · 64 GBQ8_0Runs great19 GB18
M1 Ultra · 64 GBQ8_0Runs great19 GB30
M1 Ultra · 128 GBQ8_0Runs great19 GB30

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

Information

Maker
Ai2
Released
November 2024
Parameters
8B
Longest chat
128K tokens
Reads pictures
No
Licence
llama3.1
Downloads this month
2,890
Likes
4
Sizes
4, 4.3 to 8.5 GB
Kind
Maker's release

Badge for the model card

Shows on Llama 3.1 Tulu 3 8B'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