froggeric

froggeric's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in froggeric/Qwen-Fixed-Chat-Templates about 6 hours ago

v13 to v15 are completely broken in VLLM for qwen3_coder parser

3

#16 opened about 6 hours ago by

gudsjwixvfasptjhtr

New activity in froggeric/Qwen-Fixed-Chat-Templates about 7 hours ago

v15 released with fixes for overthinking/indecision and loops on tool errors

❤️ 3

3

#15 opened about 11 hours ago by

froggeric

New activity in froggeric/Qwen-Fixed-Chat-Templates about 12 hours ago

v11/v12 performance considerations with Claude Code?

3

#11 opened 3 days ago by

mancub

test v12 + qwen 27B + llamacpp + pi coding agent

2

#12 opened 3 days ago by

tuanzxcv

Abruptly stopping with <tool call>

2

#13 opened 2 days ago by

crotron

v13 stops dead after the first response

👍 2

3

#14 opened 2 days ago by

mancub

New activity in froggeric/Qwen-Fixed-Chat-Templates 3 days ago

Template V10 causes hallucinations.

5

#10 opened 4 days ago by

lorenzodimauro97

New activity in froggeric/Qwen-Fixed-Chat-Templates 4 days ago

When using Claude Code, tool calls end up broken with this chat template in Qwen3.6-27B

6

#6 opened 5 days ago by

mancub

Qwen 3.6 template has unnecessary cache invalidations

14

#1 opened 15 days ago by

sdroege

vLLM: 'Replace the chat_template string in your tokenizer_config.json with the file contents.' ... where?

4

#5 opened 6 days ago by

pathosethoslogos

think off , but answer in reasoning

1

#7 opened 5 days ago by

xd2333

Improvement (50% token reduction via the tool)

6

#8 opened 4 days ago by

Elsephire

New activity in froggeric/Qwen-Fixed-Chat-Templates 6 days ago

Gemma 4 template request

3

#4 opened 6 days ago by

OnesimusTheLesser

New activity in froggeric/Qwen3.6-27B-MTP-GGUF 6 days ago

Thank You!

👍 1

1

#1 opened 8 days ago by

svarognsa

MTP output issues on Vulkan (6900 XT)

3

#2 opened 7 days ago by

SenorCiki

--no-mmproj-offload to avoid VRAM usage

🤝 1

2

#3 opened 7 days ago by

gdevenyi

Slow response

👍 1

3

#4 opened 6 days ago by

DarkAndBlue

New activity in froggeric/Qwen-Fixed-Chat-Templates 8 days ago

type in llama.cpp-call

#3 opened 8 days ago by

burningfeet

New activity in froggeric/Mistral-Medium-3.5-128B-MLX-4bit 11 days ago

Possibly messsed up conversion

3

#1 opened 13 days ago by

finding1

New activity in froggeric/Qwen3.6-27B-Uncensored-Heretic-v2-MLX-4bit 12 days ago

Any chance you can upload oQ4 or oQ4e mlx versions?

2

#2 opened 13 days ago by

realperson1234