Any-to-Any
MLX
Safetensors
gemma4
mlx-vlm
rlcd
multimodal
classification
parallel-inference
image-text-to-text
audio
video
4-bit precision
Instructions to use larkooo/gemma-e2b-rlcd with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use larkooo/gemma-e2b-rlcd with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download larkooo/gemma-e2b-rlcd --local-dir gemma-e2b-rlcd
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download reports/cache-trimmed.json from larkooo/gemma-e2b-rlcd: direct link, hf CLI and curl.
- Browser
- Download file 7.32 kB
-
https://huggingface.co/larkooo/gemma-e2b-rlcd/resolve/main/reports/cache-trimmed.json
- Command line
-
hf download hf://larkooo/gemma-e2b-rlcd/reports/cache-trimmed.json
-
curl -L -o cache-trimmed.json https://huggingface.co/larkooo/gemma-e2b-rlcd/resolve/main/reports/cache-trimmed.json
7.32 kB
| { | |
| "model": "models/gemma-4-e2b-it-4bit", | |
| "compute_dtype": "float32", | |
| "status": "cache_numerics_and_timing_probe_not_quality_benchmark", | |
| "results": [ | |
| { | |
| "state": "text", | |
| "prefix_tokens": 50, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.032517433166503906, | |
| "max_probability_delta": 4.89707640993764e-07, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.056962013244628906, | |
| "max_probability_delta": 5.000832680179812e-07, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 0.27352558402344584, | |
| 0.270969499979401, | |
| 0.2627255840052385 | |
| ], | |
| "cached_batched": [ | |
| 0.2238874580070842, | |
| 0.22431362498900853, | |
| 0.22326229201280512 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 0.270969499979401, | |
| "cached_batched": 0.2238874580070842 | |
| }, | |
| "speedup": 1.2102933428759866 | |
| }, | |
| { | |
| "state": "image", | |
| "prefix_tokens": 295, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.04843616485595703, | |
| "max_probability_delta": 0.0007440930557797154, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.03708457946777344, | |
| "max_probability_delta": 0.001258392859177726, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 2.799910041008843, | |
| 2.575791625014972, | |
| 2.678586375026498 | |
| ], | |
| "cached_batched": [ | |
| 0.8003228750021663, | |
| 0.774800959014101, | |
| 0.7754046249901876 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 2.678586375026498, | |
| "cached_batched": 0.7754046249901876 | |
| }, | |
| "speedup": 3.4544369335692244 | |
| }, | |
| { | |
| "state": "speech", | |
| "prefix_tokens": 86, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.08770561218261719, | |
| "max_probability_delta": 0.013604034682185018, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.03138256072998047, | |
| "max_probability_delta": 0.002520692809385161, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 0.29824562498833984, | |
| 0.3047535420046188, | |
| 0.304851624998264 | |
| ], | |
| "cached_batched": [ | |
| 0.2058587910141796, | |
| 0.1957708330010064, | |
| 0.20528816597652622 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 0.3047535420046188, | |
| "cached_batched": 0.20528816597652622 | |
| }, | |
| "speedup": 1.4845158782287822 | |
| }, | |
| { | |
| "state": "video", | |
| "prefix_tokens": 328, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.007994651794433594, | |
| "max_probability_delta": 0.00047290616548957554, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.044986724853515625, | |
| "max_probability_delta": 0.003810715236252593, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 3.470609209005488, | |
| 4.113289792003343, | |
| 3.7244844590022694 | |
| ], | |
| "cached_batched": [ | |
| 1.141944707982475, | |
| 1.2892172499850858, | |
| 1.1141800410114229 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 3.7244844590022694, | |
| "cached_batched": 1.141944707982475 | |
| }, | |
| "speedup": 3.2615278419061844 | |
| }, | |
| { | |
| "state": "video_speech", | |
| "prefix_tokens": 431, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.05509662628173828, | |
| "max_probability_delta": 0.005538062741424418, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.03436756134033203, | |
| "max_probability_delta": 0.0013212921945262024, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 4.29815650000819, | |
| 3.8582746669999324, | |
| 3.3546597079839557 | |
| ], | |
| "cached_batched": [ | |
| 1.3747120829939377, | |
| 1.1818989999883343, | |
| 1.4414640410104766 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 3.8582746669999324, | |
| "cached_batched": 1.3747120829939377 | |
| }, | |
| "speedup": 2.806605626537544 | |
| }, | |
| { | |
| "state": "long_text", | |
| "prefix_tokens": 1250, | |
| "suffix_lengths": [ | |
| 111, | |
| 99, | |
| 111, | |
| 111 | |
| ], | |
| "cached_serial_vs_uncached": { | |
| "max_logit_delta": 0.028132915496826172, | |
| "max_probability_delta": 1.1824594636777874e-09, | |
| "all_winners_match": true | |
| }, | |
| "cached_batch_vs_uncached": { | |
| "max_logit_delta": 0.0856776237487793, | |
| "max_probability_delta": 3.4840531570878852e-09, | |
| "all_winners_match": true | |
| }, | |
| "reversed_batch_vs_original": { | |
| "max_logit_delta": 0.0, | |
| "max_probability_delta": 0.0, | |
| "all_winners_match": true | |
| }, | |
| "prefix_unchanged": true, | |
| "branch_batch_sizes": [ | |
| 4 | |
| ], | |
| "warm_seconds": { | |
| "uncached": [ | |
| 3.0424842079810333, | |
| 2.5938880419998895, | |
| 2.5387540830124635 | |
| ], | |
| "cached_batched": [ | |
| 1.0249322910094634, | |
| 0.8549119170056656, | |
| 0.8327442500158213 | |
| ] | |
| }, | |
| "warm_median_seconds": { | |
| "uncached": 2.5938880419998895, | |
| "cached_batched": 0.8549119170056656 | |
| }, | |
| "speedup": 3.034099759756536 | |
| } | |
| ] | |
| } | |