hiru2005 commited on
Commit
60e7b9e
·
verified ·
1 Parent(s): 5aabe39

Model card for Sync-Strata 0.1.1 (quality, speed, support section)

Browse files
Files changed (1) hide show
  1. README.md +86 -0
README.md ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ license_name: qwen-community-1.0
4
+ license_link: LICENSE
5
+ base_model:
6
+ - Qwen/Qwen3.8-Flash-Next
7
+ base_model_relation: quantized
8
+ tags:
9
+ - qwen3.8-flash-next
10
+ - moe
11
+ - sync-strata
12
+ - experimental
13
+ ---
14
+
15
+ # Sync shared parts for Qwen3.8-Flash-Next (experimental)
16
+
17
+ The parts every Sync model (Sync-XS, Sync-S, Sync-M) needs besides its own pack. Download them once.
18
+ They are used only by **[Sync-Strata](https://github.com/hiruhiru2005/sync-strata)**; the Sync-Strata app downloads them for you.
19
+
20
+ ## Files
21
+
22
+ ### `native-gguf/`
23
+
24
+ Shared weights cut from agentionai's Gyro-S GGUF (embeddings, n-gram/PLE table and the other non-expert tensors the engine reads with `--native` / `--ple-gguf`).
25
+
26
+ The file list and sizes of this folder will be added when it is published. The whole Gyro-S file is 58.49 GB; this folder keeps only the parts the engine reads.
27
+
28
+ ### `mtp-rt/`
29
+
30
+ The multi-token-prediction (draft) layer, converted from the original BF16 checkpoint for the engine; `draft_vocab.bin` comes from upstream Strata (MIT).
31
+
32
+ Total: 0.82 GB (0.77 GiB)
33
+
34
+ | File | Size |
35
+ |---|---:|
36
+ | `dense.bin` | 116,099,072 B |
37
+ | `dense.txt` | 1,880 B |
38
+ | `draft_vocab.bin` | 425,196 B |
39
+ | `experts.bin` | 707,788,800 B |
40
+
41
+ ### `tokenizer/`
42
+
43
+ The tokenizer and chat template of Qwen3.8-Flash-Next.
44
+
45
+ Total: 0.01 GB (0.01 GiB)
46
+
47
+ | File | Size |
48
+ |---|---:|
49
+ | `chat_template.jinja` | 9,993 B |
50
+ | `merges.txt` | 3,600,844 B |
51
+ | `token_type.json` | 744,960 B |
52
+ | `tokenizer.json` | 554 B |
53
+ | `vocab.json` | 5,737,005 B |
54
+
55
+ ### `mmproj/` (optional)
56
+
57
+ Optional vision encoder and projector (BF16). **Not hosted here**: the Sync-Strata app downloads it from its original repository, [ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF](https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF). Its card says the files inherit the license of the base model (Qwen Community License 1.0); the repository metadata says Apache-2.0.
58
+
59
+ | File | Size |
60
+ |---|---:|
61
+ | `mmproj-Qwen3.8-Flash-Next-BF16.gguf` (original repository) | 907,543,008 B |
62
+
63
+ ## Where they come from
64
+
65
+ - `native-gguf` is cut from [agentionai/Qwen3.8-Flash-Next-Gyro-GGUF](https://huggingface.co/agentionai/Qwen3.8-Flash-Next-Gyro-GGUF)
66
+ (Gyro-S; the revision will be added). It keeps only the tensors the engine reads; the tensors are meant to be
67
+ byte-for-byte those of the original file, but this has not been checked with a tool yet.
68
+ - `mtp-rt` is converted from [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) (BF16) with upstream Strata's tools.
69
+ - `tokenizer` is the base model's tokenizer.
70
+ - `mmproj` is not in this repository; it is downloaded from [ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF](https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF) (converted there from the base checkpoint).
71
+
72
+ ## Credits
73
+
74
+ - Base model: [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) by the Qwen team.
75
+ - [agentionai](https://huggingface.co/agentionai) (Gyro-S), [ISTA-DASLab](https://huggingface.co/ISTA-DASLab) (vision projector),
76
+ [Strata](https://github.com/Niko1221/Strata) (MIT; tools and `draft_vocab.bin`).
77
+
78
+ ## Support (optional)
79
+
80
+ Sync-Strata and the Sync models are free. A donation page is at [Ko-fi](https://ko-fi.com/hiruhiru2005): a thank-you only, with nothing in return (no perks, no priority support, no feature promises), and it does not change the license terms.
81
+
82
+ ## License
83
+
84
+ Derived from Qwen3.8-Flash-Next: **Qwen Community License 1.0** (`LICENSE`, Copyright (c) 2026 Qwen). Keep `LICENSE`
85
+ with every copy. See the full text for the conditions on large commercial products and on Model-as-a-Service / AI work
86
+ assistant businesses.