--- title: DeepSeek-V4.1-Flash-REAP-2bit-MLX canonical_url: "https://www.modelscope.ai/models/rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX" md_url: "https://www.modelscope.ai/models/rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX.md" repository: rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX last_updated: 2026-10-15 license: mit pipeline_tag: text-generation tasks: - text-generation model_type: - deepseek_v41 architectures: - DeepseekV41ForCausalLM base_model: - deepseek-ai/DeepSeek-V4.1-Flash base_model_relation: quantized parameters: 63.9B tensor_type: - BF16 - F32 - U32 library_name: - mlx - safetensors - pytorch frameworks: - pytorch downloads: 76 stars: 0 tags: - mlx - apple-silicon - deepseek-v41 - quantized - 2-bit - reap - experimental - rapid-mlx --- # DeepSeek-V4.1-Flash-REAP-2bit-MLX > DeepSeek-V4.1-Flash-REAP-2bit-MLX - An open-source model by rapid-mlx on ModelScope. Use it from OpenCode / Claude Code / Codex: rapid-mlx agents --setup · Agent setup rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX is a 63.9B-parameter text-generation model on ModelScope. licensed under mit. derived from deepseek-ai/DeepSeek-V4.1-Flash. - **Repository**: rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX - **License**: mit - **Tasks**: text-generation - **Parameters**: 63.9B - **Base model**: deepseek-ai/DeepSeek-V4.1-Flash - **Tags**: mlx, apple-silicon, deepseek-v41, quantized, 2-bit, reap, experimental, rapid-mlx - **Downloads**: 76 - **Stars**: 0 - **Last updated**: 2026-10-15 Source: https://www.modelscope.ai/models/rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX --- ## Run it on your Mac ```bash curl -fsSL https://rapidmlx.com/install.sh | bash rapid-mlx serve deepseek-v41-flash-reap-2bit curl -s http://localhost:8000/v1/chat/completions -H 'Content-Type: application/json' -d '{"model":"default","messages":[{"role":"user","content":"Say hello"}]}' ``` Use it from OpenCode / Claude Code / Codex: `rapid-mlx agents --setup` · [Agent setup](https://rapidmlx.com/docs/agents) [Download the Desktop app](https://rapidmlx.com/download) · 256 GB unified memory required for this experimental 212.9 GB target and its DSpark sidecar. # DeepSeek V4.1 Flash · REAP · MLX 2-bit > **Experimental research artifact — not yet a product-qualified Rapid-MLX > model.** It requires a 256 GiB Apple silicon Mac and the experimental loader > that ships in [Rapid-MLX 0.14.1](https://github.com/raullenchai/Rapid-MLX/releases/tag/v0.14.1). > Standard model loaders are not supported. Vision and MTP are omitted. This is a text-only, native packed MLX artifact produced from [`Vontra/DeepSeek-V4.1-Flash-MLX-2bit-MTP`](https://huggingface.co/Vontra/DeepSeek-V4.1-Flash-MLX-2bit-MTP), itself derived from [`deepseek-ai/DeepSeek-V4.1-Flash`](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash). The conversion preserves the source affine 2-bit/group-64 values while stacking routed experts for native batched execution. Conservative routing-saliency pruning keeps 336 of 384 routed experts in every layer. ## Qualification result Hardware: Apple M3 Ultra, 60 GPU cores, 256 GiB unified memory, MLX 0.32.2. | Measurement | Result | | --- | ---: | | Tensor payload | 212,930,051,680 bytes (~199 GiB) | | Strict resident load | 239.44 seconds | | Peak MLX memory | 213.51 GB | | Conservative decode | 7.31 tok/s | | Best short-decode ceiling | 7.92 tok/s | | Routed experts retained | 336 / 384 per layer | | Mean retained routing mass | 99.9949% | | Worst-layer retained routing mass | 99.8564% | Decode throughput uses greedy generation and `(N-1)/elapsed`, excluding prefill. The same greedy token chain was reproduced at evaluation intervals 4, 20, and 40. Correctly framed smoke output was `The capital of France is Paris.` followed by EOS. The artifact misses Rapid-MLX's 12 tok/s product floor by 34.0% and is therefore published for reproducibility and further optimization only. Uploading it does not mean that Rapid-MLX catalog, server, or Desktop support is available. ## Initial evaluation results These are the initial qualification checks completed before publication. They are intentionally reported as narrow pass/fail evidence, not as a general model quality score. | Check | Result | Evidence | | --- | --- | --- | | One-sentence factual chat | Pass | Prompt: `What is the capital of France? Answer in one short sentence.` Output: `The capital of France is Paris.` then EOS. | | Greedy stability | Pass | Evaluation intervals 4, 20, and 40 produced the identical token sequence. | | Prefill/decode numerical parity | Pass | Tiny-config prefill, decode, and chunked-prefill relative differences were at most 1.4e-6. | | Next-token parity | Pass | 100% argmax agreement across the tiny-config comparison battery. | | Quantized tensor fidelity | Pass | Quantize/dequantize cases were bit-exact; strict float and quantized layouts round-tripped. | | Full artifact load | Pass | All 41 shards strict-loaded with no missing or unexpected text-model parameters. | | Vision, tool use, long context | Not evaluated | These capabilities are not claimed by this artifact. | The factual chat case used the required model framing: ```text <｜begin▁of▁sentence｜><｜User｜>What is the capital of France? Answer in one short sentence.<｜Assistant｜> ``` Raw prompts without these role markers produced invalid diagnostics and are not counted as evaluation results. ## Calibration and validation - REAP calibration used 2,048 real causal tokens split into two disjoint halves. - Split-half expert keep-set overlap was 92.1354%. - Tiny-config parity checks passed with relative differences no greater than 1.4e-6 and 100% argmax agreement. - Quantize/dequantize cases were bit-exact; strict float/quantized layouts and streaming output matched. - The 41-shard artifact strict-loaded without missing or unexpected text-model parameters. These checks do not establish broad quality, tool-use, long-context, or safety parity with the original model. Two-bit quantization and expert pruning may reduce quality. Long context has not been qualified. ## Experimental usage Clone the exact qualification implementation and download the model through the standard Hugging Face cache: ```bash git clone --branch v0.14.1 https://github.com/raullenchai/Rapid-MLX.git cd Rapid-MLX hf download rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX python scripts/qualify_deepseek_v41_native.py \ ~/.cache/huggingface/hub/models--rapid-mlx--DeepSeek-V4.1-Flash-REAP-2bit-MLX/snapshots/ \ --tokens 16 --eval-intervals 4 ``` Allow substantial memory headroom for macOS and context growth. This artifact is not suitable for machines below 256 GiB unified memory. ## Background The runtime is a native MLX implementation of the V4.1 architecture — attention, MoE, engram, hyper-connections, indexer, sparse attention and a chunked-prefill streaming generator — written for this qualification and shipped in Rapid-MLX 0.14.1 under `scripts/deepseek_v41_native/`. Why the artifact is published even though it misses the product floor, and what the numbers mean in practice, is written up here: [DeepSeek V4.1 on a Mac, and Why Your LLM API Router Is Reading Your Prompts](https://rapidmlx.com/blog/deepseek-v41-on-a-mac-and-the-router-problem). For models that *are* product-qualified and run on an ordinary Mac, see [rapidmlx.com](https://rapidmlx.com). ## License and attribution The MIT license from the original release is included in `LICENSE`. Model design and training belong to DeepSeek and its contributors. The source 2-bit conversion was published by Vontra; this independent REAP repack was produced by the Rapid-MLX project. See `NOTICE` for the pinned source revision.