MLX
Safetensors
qwen3
dflash2
draft-only
speculative-decoding
block-diffusion
draft-model
qwen3.8-27b-target
Instructions to use RESMP-DEV/DFlash2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use RESMP-DEV/DFlash2 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download RESMP-DEV/DFlash2 --local-dir DFlash2
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download metrics.json from RESMP-DEV/DFlash2: direct link, hf CLI and curl.
- Browser
- Download file 2.18 kB
-
https://huggingface.co/RESMP-DEV/DFlash2/resolve/main/metrics.json
- Command line
-
hf download hf://RESMP-DEV/DFlash2/metrics.json
-
curl -L -o metrics.json https://huggingface.co/RESMP-DEV/DFlash2/resolve/main/metrics.json
2.18 kB
| { | |
| "schema_version": 1, | |
| "decision": "promoted", | |
| "model_sha256": "c90e0fc65dddbe4107b1b417d771c3be188c87197e89cf4b14415f56abc866f0", | |
| "config_sha256": "873e3556509b0da06e29654ba00d4944888d4b5e8a33afde25f7eb27d321e980", | |
| "blend_sha256": "e53a6bb5e890d7e3368053302189766c3ef68d2eb7887152ece1ea9eac4f64d5", | |
| "blend": { | |
| "generation_2_strength": 0.0625, | |
| "generation_3_strength": 0.1875 | |
| }, | |
| "runtime": { | |
| "implementation": "DFlash MLX", | |
| "revision": "dcf93d3d07aadc5a2e74fab7e44c63209d9f37a3", | |
| "hardware": "Apple M4 Max", | |
| "target": "calibrated Qwen3.8-27B MXFP4 runtime", | |
| "draft_quantization": "none" | |
| }, | |
| "exact_parity": { | |
| "token_match": true, | |
| "acceptance_ratio": 0.7109375, | |
| "baseline_tps_diagnostic": 21.961506855085254, | |
| "dflash_tps_diagnostic": 33.51110051912492 | |
| }, | |
| "paired_acceptance": [ | |
| { | |
| "gate": "legacy24", | |
| "control_mean_acceptance": 0.6712239583333334, | |
| "candidate_mean_acceptance": 0.671875, | |
| "mean_delta": 0.0006510416666666297, | |
| "candidate_wins": 5, | |
| "ties": 14, | |
| "losses": 5 | |
| }, | |
| { | |
| "gate": "fable-heldout25", | |
| "control_mean_acceptance": 0.6471875, | |
| "candidate_mean_acceptance": 0.64921875, | |
| "mean_delta": 0.002031249999999929, | |
| "candidate_wins": 11, | |
| "ties": 5, | |
| "losses": 9 | |
| } | |
| ], | |
| "training_capture": { | |
| "cycle_count": 11273, | |
| "prompt_group_count": 127, | |
| "legacy_cycles": 1981, | |
| "legacy_prompts": 24, | |
| "fable_training_cycles": 9292, | |
| "fable_training_prompts": 103, | |
| "fable_heldout_prompts_excluded": 25, | |
| "teacher_target_topk": 64, | |
| "teacher_precision": "calibrated MXFP4 target runtime" | |
| }, | |
| "training_metrics": { | |
| "best_validation_prefix_accuracy": 0.793221252777784, | |
| "best_validation_top1_accuracy": 0.871487798557914 | |
| }, | |
| "non_claims": [ | |
| "Acceptance metrics do not prove broad task quality or natural-EOS behavior.", | |
| "Throughput and memory are diagnostic because thermal pressure was not controlled.", | |
| "The Fable held-out prompts come from the same eval shard as Fable training prompts.", | |
| "No trace data or capture cycles are included in this model repository." | |
| ] | |
| } | |