Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
caiovicentino1
/
Qwopus3.5-9B-v3-HLWQ-Q5
like
9
Text Generation
Safetensors
qwen3_5
hlwq
gptq
int4
qwen3.5
vllm
marlin
conversational
Eval Results (legacy)
4-bit precision
arxiv:
2502.02617
arxiv:
2603.29078
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
3
Use this model
main
Qwopus3.5-9B-v3-HLWQ-Q5
16.3 GB
Ctrl+K
Ctrl+K
1 contributor
History:
45 commits
caiovicentino1
Remove legacy polar_config.json
fe18601
verified
9 days ago
.gitattributes
Safe
1.67 kB
add: benchmark charts (HumanEval, speed, size)
15 days ago
Jackrong_Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled-v2_humaneval_all_evalonly_eval_results.json
Safe
142 kB
PolarQuant Q5 unified (PPL 6.48, 7.1GB, 42 tok/s)
22 days ago
Jackrong_Qwopus3.5-9B-test1_humaneval_all_evalonly_eval_results.json
Safe
176 kB
PolarQuant Q5 unified (PPL 6.48, 7.1GB, 42 tok/s)
22 days ago
README.md
Safe
6.33 kB
HLWQ rebrand: title, tags, notice, self-links
9 days ago
benchmarks.png
Safe
119 kB
xet
add: benchmark charts (HumanEval, speed, size)
15 days ago
chat_template.jinja
Safe
4.05 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
config.json
Safe
4.23 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
generation_config.json
Safe
225 Bytes
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
hlwq_config.json
49 kB
Add hlwq_config.json (rename from polar_config.json)
9 days ago
kv_benchmark.png
Safe
82.3 kB
add kv_benchmark.png
16 days ago
kv_context.png
Safe
110 kB
xet
add kv_context.png
16 days ago
model-00001-of-00003.safetensors
Safe
2.96 GB
xet
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
model-00002-of-00003.safetensors
Safe
4.29 GB
xet
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
model-00003-of-00003.safetensors
Safe
1.34 GB
xet
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
model.safetensors
Safe
7.65 GB
xet
fix: unpack lm_head + embed_tokens + in_proj_a/b to BF16 (Marlin kernel compat)
16 days ago
model.safetensors.index.json
Safe
128 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
ppl_comparison.png
Safe
45.6 kB
Fix chart: ppl_comparison.png with correct 9B values
22 days ago
processor_config.json
Safe
1.3 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
quant_log.csv
Safe
9.61 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
quantize_config.json
Safe
1.04 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
qwen_Qwen3.5-9B_humaneval_all_evalonly_eval_results.json
Safe
135 kB
PolarQuant Q5 unified (PPL 6.48, 7.1GB, 42 tok/s)
22 days ago
speed_vram.png
Safe
52.3 kB
Fix chart: speed_vram.png with correct 9B values
22 days ago
tokenizer.json
Safe
20 MB
xet
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
tokenizer_config.json
Safe
1.17 kB
upgrade: GPTQ calibrated INT4 W4A16 desc_act=True (GPTQModel 6.0.3, 512 samples, HumanEval 60.98%)
15 days ago
