Hugging FaceA 177B, 59B active chat model from agentionai that reads pictures as well as text.
- Downloads
- 749K
- Likes
- 166
- Parameters
- 177B, 59B active
- Context
- 256K
- Licence
- other
Too big for this Mac
Its smallest size needs 60 GB of memory, more than M4 · 16 GB can give it.
Qwen3.8 Flash Next Gyro runs well from a Mac with an M2 Max and 96 GB of memory. 15 of 59 Mac configurations run it well.
Compare with
The nearest models in size.
- Swift 1.5 Qwen3.8 Flash Next GSQ RCOukisai · 771K downloadsToo big for this Mac · 67 GB
- POCKET Darwin 180BFINAL-Bench · 403K downloadsToo big for this Mac · 111 GB
- Swift 1.5 Qwen3.8 Flash Nextukisai · 160K downloadsToo big for this Mac · 70 GB
- Qwen3.8 Flash NextAlibaba Qwen · 9.4M downloadsToo big for this Mac · 66 GB
Sizes
From agentionai/Qwen3.8-Flash-Next-Gyro-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.
| Size | Download | Memory at 8K | Get |
|---|---|---|---|
| TQ1_0 | 58 GB | 59 GBestimated | Open |
| TQ2_0 | 92 GB | 92 GBestimated | Open |
| TQ3_0 | 108 GB | 108 GBestimated | Open |
Run It Anywhere
Qwen3.8 Flash Next Gyro is an open model, so any app that runs GGUF files can run it. These take the TQ1_0 size from agentionai straight from Hugging Face.
- LM StudioMac, Windows, Linux
Open - JanMac, Windows, Linux
Open - MstyMac, Windows, Linux
Open - OllamaMac, Windows, Linux
ollama run hf.co/agentionai/Qwen3.8-Flash-Next-Gyro-GGUF:TQ1_0 - llama.cppMac, Windows, Linux
llama-server -hf agentionai/Qwen3.8-Flash-Next-Gyro-GGUF:TQ1_0 - Docker Model RunnerMac, Windows, Linux
docker model run hf.co/agentionai/Qwen3.8-Flash-Next-Gyro-GGUF:TQ1_0
On a PC Graphics Card
| Video memory | Largest size that fits | Download | Memory at 8K |
|---|---|---|---|
| 6 GB | None worth running | needs 108 GB | |
| 8 GB | None worth running | needs 108 GB | |
| 10 GB | None worth running | needs 108 GB | |
| 11 GB | None worth running | needs 108 GB | |
| 12 GB | None worth running | needs 108 GB | |
| 16 GB | None worth running | needs 108 GB | |
| 20 GB | None worth running | needs 108 GB | |
| 24 GB | None worth running | needs 108 GB | |
| 32 GB | None worth running | needs 108 GB | |
| 48 GB | None worth running | needs 108 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
| Mac | Size | Fit | Memory | Writes |
|---|---|---|---|---|
| M6 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M6 · 24 GB | TQ1_0 | Too big | 60 GB | |
| M6 · 32 GB | TQ1_0 | Too big | 60 GB | |
| M5 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M5 · 24 GB | TQ1_0 | Too big | 60 GB | |
| M5 · 32 GB | TQ1_0 | Too big | 60 GB | |
| M5 Pro · 24 GB | TQ1_0 | Too big | 60 GB | |
| M5 Pro · 48 GB | TQ1_0 | Too big | 60 GB | |
| M5 Pro · 64 GB | TQ1_0 | Too big | 60 GB | |
| M5 Max · 36 GB | TQ1_0 | Too big | 60 GB | |
| M5 Max · 48 GB | TQ1_0 | Too big | 60 GB | |
| M5 Max · 64 GB | TQ1_0 | Too big | 60 GB | |
| M5 Max · 128 GB | TQ2_0 | Runs well | 93 GB | 8.8 |
| M5 Ultra · 96 GB | TQ1_0 | Runs well | 60 GB | 26 |
| M5 Ultra · 256 GB | TQ3_0 | Runs great | 109 GB | 15 |
| M5 Ultra · 512 GB | TQ3_0 | Runs great | 109 GB | 15 |
| M4 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M4 · 24 GB | TQ1_0 | Too big | 60 GB | |
| M4 · 32 GB | TQ1_0 | Too big | 60 GB | |
| M4 Pro · 24 GB | TQ1_0 | Too big | 60 GB | |
| M4 Pro · 48 GB | TQ1_0 | Too big | 60 GB | |
| M4 Pro · 64 GB | TQ1_0 | Too big | 60 GB | |
| M4 Max · 36 GB | TQ1_0 | Too big | 60 GB | |
| M4 Max · 48 GB | TQ1_0 | Too big | 60 GB | |
| M4 Max · 64 GB | TQ1_0 | Too big | 60 GB | |
| M4 Max · 96 GB | TQ1_0 | Runs well | 60 GB | 8.7 |
| M4 Max · 128 GB | TQ2_0 | Runs well | 93 GB | 7.5 |
| M3 · 8 GB | TQ1_0 | Too big | 60 GB | |
| M3 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M3 · 24 GB | TQ1_0 | Too big | 60 GB | |
| M3 Pro · 18 GB | TQ1_0 | Too big | 60 GB | |
| M3 Pro · 36 GB | TQ1_0 | Too big | 60 GB | |
| M3 Max · 36 GB | TQ1_0 | Too big | 60 GB | |
| M3 Max · 48 GB | TQ1_0 | Too big | 60 GB | |
| M3 Max · 64 GB | TQ1_0 | Too big | 60 GB | |
| M3 Max · 96 GB | TQ1_0 | Runs well | 60 GB | 6.7 |
| M3 Max · 128 GB | TQ2_0 | Runs well | 93 GB | 5.7 |
| M3 Ultra · 96 GB | TQ1_0 | Runs well | 60 GB | 16 |
| M3 Ultra · 256 GB | TQ3_0 | Runs great | 109 GB | 9.3 |
| M3 Ultra · 512 GB | TQ3_0 | Runs great | 109 GB | 9.3 |
| M2 · 8 GB | TQ1_0 | Too big | 60 GB | |
| M2 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M2 · 24 GB | TQ1_0 | Too big | 60 GB | |
| M2 Pro · 16 GB | TQ1_0 | Too big | 60 GB | |
| M2 Pro · 32 GB | TQ1_0 | Too big | 60 GB | |
| M2 Max · 32 GB | TQ1_0 | Too big | 60 GB | |
| M2 Max · 64 GB | TQ1_0 | Too big | 60 GB | |
| M2 Max · 96 GB | TQ1_0 | Runs well | 60 GB | 8.6 |
| M2 Ultra · 64 GB | TQ1_0 | Too big | 60 GB | |
| M2 Ultra · 128 GB | TQ2_0 | Runs well | 93 GB | 11 |
| M2 Ultra · 192 GB | TQ3_0 | Runs great | 109 GB | 9.3 |
| M1 · 8 GB | TQ1_0 | Too big | 60 GB | |
| M1 · 16 GB | TQ1_0 | Too big | 60 GB | |
| M1 Pro · 16 GB | TQ1_0 | Too big | 60 GB | |
| M1 Pro · 32 GB | TQ1_0 | Too big | 60 GB | |
| M1 Max · 32 GB | TQ1_0 | Too big | 60 GB | |
| M1 Max · 64 GB | TQ1_0 | Too big | 60 GB | |
| M1 Ultra · 64 GB | TQ1_0 | Too big | 60 GB | |
| M1 Ultra · 128 GB | TQ2_0 | Runs well | 93 GB | 9.9 |
Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.
Information
- Maker
- agentionai
- Released
- September 2026
- Parameters
- 177B, 59B active
- Longest chat
- 256K tokens
- Reads pictures
- Yes
- Licence
- other
- Downloads this month
- 748,659
- Likes
- 166
- Sizes
- 3, 58 to 108 GB
- Kind
- Maker's release
Badge for the model card
Shows on Qwen3.8 Flash Next Gyro's Hugging Face card which Macs run it well, linked to this page. It updates every morning.
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.