Hugging Face
Search models, datasets, users...
Ctrl K
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
TelperionAI
/
Qwen3.8-27B-MLA
Like
18
Follow
TelperionAI
32
Text Generation
Safetensors
qwen3_5
multi-head-latent-attention
mla
kv-cache-compression
long-context
qwen3
conversational
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
1
Copy to bucket
main
Qwen3.8-27B-MLA
55.7 GB
Ctrl+K
Ctrl+K
2 contributors
History:
6 commits
QKV768
Calibrated FP8 KV-cache scales (mla_kv_cache_scales); card: family, FP8 KV, MTP caveat. Needs qwen-mla-vllm >= 0.5.0
572038a
verified
about 1 month ago
.gitattributes
1.57 kB
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
README.md
14 kB
Calibrated FP8 KV-cache scales (mla_kv_cache_scales); card: family, FP8 KV, MTP caveat. Needs qwen-mla-vllm >= 0.5.0
about 1 month ago
chat_template.jinja
8.95 kB
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
config.json
5.72 kB
Calibrated FP8 KV-cache scales (mla_kv_cache_scales); card: family, FP8 KV, MTP caveat. Needs qwen-mla-vllm >= 0.5.0
about 1 month ago
generation_config.json
202 Bytes
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
merges.txt
3.35 MB
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
model-00001-of-00002.safetensors
49.9 GB
xet
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
model-00002-of-00002.safetensors
3.98 GB
xet
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
model-mtp.safetensors
849 MB
xet
Add the base model's vision tower and MTP head (Qwen3_5MLAForConditionalGeneration); card: vision + MTP results. Needs qwen-mla-vllm >= 0.4.0
about 2 months ago
model-visual.safetensors
921 MB
xet
Add the base model's vision tower and MTP head (Qwen3_5MLAForConditionalGeneration); card: vision + MTP results. Needs qwen-mla-vllm >= 0.4.0
about 2 months ago
model.safetensors.index.json
113 kB
Add the base model's vision tower and MTP head (Qwen3_5MLAForConditionalGeneration); card: vision + MTP results. Needs qwen-mla-vllm >= 0.4.0
about 2 months ago
preprocessor_config.json
390 Bytes
Add the base model's vision tower and MTP head (Qwen3_5MLAForConditionalGeneration); card: vision + MTP results. Needs qwen-mla-vllm >= 0.4.0
about 2 months ago
tokenizer.json
12.8 MB
xet
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
tokenizer_config.json
17.9 kB
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago
video_preprocessor_config.json
385 Bytes
Add the base model's vision tower and MTP head (Qwen3_5MLAForConditionalGeneration); card: vision + MTP results. Needs qwen-mla-vllm >= 0.4.0
about 2 months ago
vocab.json
6.72 MB
Qwen3.8-27B-MLA v1 preview: training-free MLA retrofit, 2x KV, bf16 weights
about 2 months ago