Publish DFlash pairing index d3c66d8d895b504538a0c5f3e75caef1b95324543aaeb98a628feff9464d1e11 from mesh-llm 2f83990320e0109c0529f26297554606f8c0b5e4
c7c49ad verified | license: apache-2.0 | |
| tags: [mesh-llm, dflash, speculative-decoding] | |
| # DFlash pairing index for mesh-llm | |
| Each pairing maps a target model's GGUF identity (architecture, executable layer count, width, expert | |
| count, optional family) to the DFlash draft mesh-llm uses by default. The draft is pinned by Hugging | |
| Face revision and SHA-256, and the pairing carries the default draft cap and the backends it is the | |
| default on. | |
| Every pairing here was chosen from measured decode speed on DGX Spark (CUDA) and Apple Silicon | |
| (Metal). A pairing is the default on a backend only where it beat the model's previous default: | |
| native MTP where the model ships it, else no speculation. | |
| ## Files | |
| - `index/<sha256>.json`: every index a mesh-llm release has pinned, named by the SHA-256 of its bytes. | |
| A release fetches the file its pin names from `main` and uses it only if the bytes hash to the pin. | |
| These files are never changed or deleted, so old releases keep resolving. | |
| - `index.json`: a copy of the newest index, for reading. No release reads it. | |
| The source of truth is | |
| [`mesh/crates/mesh-llm-skippy-adapter/data/dflash-pairings/`](https://github.com/Mesh-LLM/mesh-llm/tree/main/mesh/crates/mesh-llm-skippy-adapter/data/dflash-pairings) | |
| in mesh-llm. To change the pairings, edit `index.json` there, set `DFLASH_PAIRING_INDEX_SHA256` to | |
| its new hash, and merge. CI then publishes `index/<sha256>.json`, `index.json` and this README. | |
| ## Pairings | |
| <!-- pairings:start --> | |
| | Match | Draft | Cap | Backends | | |
| | --- | --- | --- | --- | | |
| | `qwen35`, 32 layers, 4096 wide | [AtomicChat/Qwen3.5-9B-DFlash-GGUF](https://huggingface.co/AtomicChat/Qwen3.5-9B-DFlash-GGUF/blob/c7c69923d77a9aa0004b410d3a7a3dbb64591d41/Qwen3.5-9B-DFlash.Q8_0.gguf) `Qwen3.5-9B-DFlash.Q8_0.gguf` | 15 | all | | |
| | `qwen35`, 64 layers, 5120 wide | [AtomicChat/Qwen3.6-27B-DFlash-GGUF](https://huggingface.co/AtomicChat/Qwen3.6-27B-DFlash-GGUF/blob/a3d4fb1eff6f268cbbdf808042f6902cc17f5458/Qwen3.6-27B-DFlash.Q8_0.gguf) `Qwen3.6-27B-DFlash.Q8_0.gguf` | 15 | all | | |
| | `qwen35`, 64 layers, 5120 wide, name carries `Qwen3.8` | [z-lab/Qwen3.8-27B-DFlash2-GGUF](https://huggingface.co/z-lab/Qwen3.8-27B-DFlash2-GGUF/blob/2d9571f8ce46e151f61c6499c99dee6079e1d610/Qwen3.8-27B-DFlash2-Q8_0.gguf) `Qwen3.8-27B-DFlash2-Q8_0.gguf` | 7 | all | | |
| | `qwen35moe`, 40 layers, 2048 wide, 256 experts | [aminya/Qwen3.6-35B-A3B-DFlash2-GGUF](https://huggingface.co/aminya/Qwen3.6-35B-A3B-DFlash2-GGUF/blob/1e26e3499d865b9d2fc55e019d65b3f0e8ab6846/Qwen3.6-35B-A3B-DFlash2-Q8_0.gguf) `Qwen3.6-35B-A3B-DFlash2-Q8_0.gguf` | 7 | not Metal | | |
| | `qwen35moe`, 40 layers, 2048 wide, 256 experts | [Alittlehammmer/Qwen3.6-35B-A3B-DFlash-GGUF-llama.cpp](https://huggingface.co/Alittlehammmer/Qwen3.6-35B-A3B-DFlash-GGUF-llama.cpp/blob/2aa57b642f211d2639083954c5b66d9c1ee503fc/Qwen3.6-35B-A3B-DFlash-Q8_0.gguf) `Qwen3.6-35B-A3B-DFlash-Q8_0.gguf` | 15 | Metal | | |
| | `qwen3next`, 48 layers, 2048 wide, 512 experts | [AtomicChat/Qwen3-Coder-Next-DFlash-GGUF](https://huggingface.co/AtomicChat/Qwen3-Coder-Next-DFlash-GGUF/blob/f5b4c9d2511678f75a41b3d3de5d9e379dd88768/Qwen3-Coder-Next-DFlash.Q8_0.gguf) `Qwen3-Coder-Next-DFlash.Q8_0.gguf` | 15 | Metal | | |
| | `gemma4`, 42 layers, 2560 wide, name carries `Nemotron-3.5-Bolt` | [unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF](https://huggingface.co/unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF/blob/b738967d6da3441a7d1c71fcec15e49e32e16ecd/dflash-NVIDIA-Nemotron-3.5-Bolt-E4B-Q8_0.gguf) `dflash-NVIDIA-Nemotron-3.5-Bolt-E4B-Q8_0.gguf` | 7 | all | | |
| | `nemotron_h_moe`, 52 layers, 2688 wide, 128 experts, name carries `Nemotron-3.5-Lightning` | [magnitudedev/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4-DSpark-GGUF](https://huggingface.co/magnitudedev/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4-DSpark-GGUF/blob/ed35fc990759330e325ab9e41bcf80737ab4dafc/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4-DSpark.gguf) `NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4-DSpark.gguf` | 7 | not Metal | | |
| | `laguna`, 48 layers, 3072 wide, 256 experts | [poolside/Laguna-S-2.1-GGUF](https://huggingface.co/poolside/Laguna-S-2.1-GGUF/blob/19036076775e9ba6758595f78af99cb976ef8ff0/laguna-s-2.1-DFlash-BF16.gguf) `laguna-s-2.1-DFlash-BF16.gguf` | 15 | not Metal | | |
| | `deepseek4`, 43 layers, 4096 wide, 256 experts, name carries `DeepSeek-V4-Flash-0731` | [unsloth/DeepSeek-V4-Flash-0731-GGUF](https://huggingface.co/unsloth/DeepSeek-V4-Flash-0731-GGUF/blob/fbbb5b93fb787c21338159b0af3318bb3f4d9768/dspark-DeepSeek-V4-Flash-0731-Q8_0.gguf) `dspark-DeepSeek-V4-Flash-0731-Q8_0.gguf` | 5 | not Metal | | |
| <!-- pairings:end --> | |
| The cap is further held to 6 on Metal. Layer packages can instead declare their draft in | |
| `generation.speculative_decoding.proposers` (`type: dflash`); a package declaration takes precedence | |
| over this index. | |