moyanwang
bigmoyan
AI & ML interests
AIGC / LLM / Large Multimodal Model/ Infra
Organizations
use-fast-tokenizer-v2
👍 1
#40 opened 2 months ago
by
bigmoyan
Fast tokenizer token-ID parity with previous slow tokenizer?
3
#39 opened 2 months ago
by
bullpoint
use-fast-tokenizer
1
#38 opened 2 months ago
by
bigmoyan
support-transformers-inference
2
#115 opened 3 months ago
by
bigmoyan
fix: add missing use_deterministic_attn parameter to MoonViT3dEncoder
5
#22 opened 3 months ago
by
ace-coreweave
support-transformers-inference
#26 opened 3 months ago
by
bigmoyan
AttributeError: 'CompressedLinear' object has no attribute 'weight'
2
#39 opened 5 months ago
by
xbeark
Add tokenizer.json for Kimi K2.5
3
#108 opened 3 months ago
by
hoangquan456
Chat Template Logic Issue: Ambiguous Default Thinking Mode
6
#27 opened 6 months ago
by
QIN2DIM
fix: allow pass image kwargs to image processor
3
#90 opened 4 months ago
by
bigmoyan
Remove default system prompt, fix media_start token and update readme
#20 opened 6 months ago
by
bigmoyan
Compatibility with transformers v5
2
#11 opened 6 months ago
by
pcuenq
Rendered text for moonshotai/Kimi-K2-Thinking seems not correct.
1
#3 opened 8 months ago
by
bigmoyan
fix-default-system-prompt
🚀❤️ 2
2
#12 opened 8 months ago
by
bigmoyan
fix vocab size
1
#4 opened 8 months ago
by
bigmoyan
fix-vocab-size
1
#17 opened 8 months ago
by
bigmoyan
Adjust number of reserved tokens to match the model
1
#15 opened about 1 year ago
by
dzhulgakov
Update tokenization_kimi.py
1
#56 opened 11 months ago
by
lfu
remove-added-functions.-in-tool_call_id
1
#60 opened 9 months ago
by
bigmoyan
not add functions. in tool id
1
#15 opened 9 months ago
by
bigmoyan