Spaces:

RonitP1
/

KhanaVision

Sleeping

App Files Files Community

97.6 kB

Ctrl+K

Ctrl+K

1 contributor

History: 10 commits

Ronti Patange

Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4

e5ffef0 30 days ago

.gitignore

257 Bytes
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
BENCHMARK.md

26.7 kB
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
README.md

377 Bytes
Fix invalid gradio sdk_version, update README for Qwen3-VL-4B about 1 month ago
app.py

4.42 kB
Add RAW_PROMPT: mode to analyze() for direct benchmark comparison about 1 month ago
bench_fewshot.py

3.79 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_gemma.py

5.12 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_hf.py

5.52 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_kaggle80.py

3.69 kB
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
bench_kaggle80_v2.py

4.14 kB
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
bench_llamacpp.py

3.34 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_llamacpp_v2.py

5.02 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_moondream.py

4.24 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_parallel.py

4.98 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_parallel_qwen3.py

4.98 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_qwen3.py

5.01 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago
bench_zerogpu.py

5.23 kB
Read KHANA_API_KEY from env var instead of hardcoding it 30 days ago
kaggle_classes.txt

889 Bytes
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
kaggle_classes_desc.py

7.42 kB
Add Kaggle 80-class hard-mode benchmark for Qwen3-VL-4B q4 30 days ago
requirements.txt

64 Bytes
Add torchvision (required by Qwen3-VL's video sub-processor) about 1 month ago
test_ollama.py

2.42 kB
Add full VLM benchmark suite and results: model comparison, prompt engineering, ZeroGPU deploy 30 days ago