hmkim97
/

tangram-gate

kv-cache-compression

Model card Files Files and versions

Commit History

Upload qwen3-30b-a3b-instruct-2507/q8_dim16_sink16.pt with huggingface_hub

06a628f
verified

hmkim97 commited on 23 days ago

List gpt-oss-20b gate

a235358
verified

Kim commited on May 30

Add gpt-oss-20b gate (q8_dim16_sink16)

34b07d4
verified

Kim commited on May 30

Soften wording: single Trained gates table, drop mirror disclaimer

c290d8c
verified

Kim commited on May 30

Remove arXiv link to drop auto paper association

739b4cd
verified

Kim commited on May 30

Upload gemma-3-12b-it/q2_dim16_sink16.pt with huggingface_hub

a9fe8df
verified

Kim commited on May 30

Upload qwen2.5-7b-instruct-1m/q7_dim16_sink16.pt with huggingface_hub

4f927d3
verified

Kim commited on May 30

Upload qwen3-4b-instruct-2507/q4_dim16_sink16.pt with huggingface_hub

fe91d37
verified

Kim commited on May 30

Upload qwen3-8b/q4_dim16_sink16.pt with huggingface_hub

b12b1e2
verified

Kim commited on May 30

Upload qwen3-14b/q5_dim16_sink16.pt with huggingface_hub

3193c27
verified

Kim commited on May 30

Upload llama3.1-8b-instruct/q4_dim16_sink16_v0.pt with huggingface_hub

afb213c
verified

Kim commited on May 30

Upload llama3.1-8b-instruct/q4_dim16_sink16.pt with huggingface_hub

75dabb0
verified

Kim commited on May 30

Upload README.md with huggingface_hub

10c29e2
verified

Kim commited on May 30

initial commit

f0ca9b2
verified

Kim commited on May 30