モデル情報

DeepSeek-V4.1-Flash-ROCMFP23-GGUF

Lucebox · DeepSeek-V4.1-Flash-ROCMFP23-GGUF

Lucebox builds of DeepSeek V4.1 Flash for the Lucebox DS4.1 engine (AMD R9700 + Strix Halo). Routed experts in the engine's ROCMFP block formats; attention, shared experts and head in Q8_0; both Engram tables embedded. Calibrated with the published 262,144-token imatrix (`smalinin/DeepSeek-V4.1-Flash-GGUF`, converted in `imatrix/`). Formats: `Q2_0_ROCMFP2` (ggml type 107, 10 bytes per 32 weights, 2.5 bits, codebook {-1, 0, 1, 2} x a ue4m3 scale per 16 weights), `Q3_0_ROCMFPX` (type 104, 14 bytes per 32, 3.5 bits). **fp2s** files use type 107 with bit 7 of a half-block's scale byte meaning "mirror the codebook"; a reader that ignores that bit produces garbage, so use an engine build that honors it. KL is measured on 8,184 held-out fineweb-edu tokens against the unquantized model through the same reference forward pass (`eval/` holds the teacher log-probs and the tokens). Recipe, converter patches and harness: Lucebox repo, `lucebox_training/quantization`. `DeepSeek-V4.1-Flash-mmproj-BF16.gguf` (0.90 GiB): this checkpoint's vision encoder (32-layer ViT and the aligner into the 5120-wide model), its 266 tensors copied byte for byte in BF16 by the engine's `server/tools/export_ds4v_mmproj.py`, with the checkpoint revision recorded (sha256 in `vision/`). Pass it with `--mmproj` to an engine build with V4.1 image input, next to these September files only: the October 17 checkpoint retrained its encoder and ships its own in `Lucebox/DeepSeek-V4.1-Flash-1017-ROCMFP-GGUF`. The engine refuses an encoder from another checkpoint than the target's.

翻訳待ちのため原文を表示しています。

license:mit
コンテキスト長非公開
入力 → 出力待核验 → 待核验
モデルウェイト未確認
情報更新日時2026/11/2 19:15:32
OVERVIEW

モデル紹介

Lucebox builds of DeepSeek V4.1 Flash for the Lucebox DS4.1 engine (AMD R9700 + Strix Halo). Routed experts in the engine's ROCMFP block formats; attention, shared experts and head in Q8_0; both Engram tables embedded. Calibrated with the published 262,144-token imatrix (`smalinin/DeepSeek-V4.1-Flash-GGUF`, converted in `imatrix/`). Formats: `Q2_0_ROCMFP2` (ggml type 107, 10 bytes per 32 weights, 2.5 bits, codebook {-1, 0, 1, 2} x a ue4m3 scale per 16 weights), `Q3_0_ROCMFPX` (type 104, 14 bytes per 32, 3.5 bits). **fp2s** files use type 107 with bit 7 of a half-block's scale byte meaning "mirror the codebook"; a reader that ignores that bit produces garbage, so use an engine build that honors it. KL is measured on 8,184 held-out fineweb-edu tokens against the unquantized model through the same reference forward pass (`eval/` holds the teacher log-probs and the tokens). Recipe, converter patches and harness: Lucebox repo, `lucebox_training/quantization`. `DeepSeek-V4.1-Flash-mmproj-BF16.gguf` (0.90 GiB): this checkpoint's vision encoder (32-layer ViT and the aligner into the 5120-wide model), its 266 tensors copied byte for byte in BF16 by the engine's `server/tools/export_ds4v_mmproj.py`, with the checkpoint revision recorded (sha256 in `vision/`). Pass it with `--mmproj` to an engine build with V4.1 image input, next to these September files only: the October 17 checkpoint retrained its encoder and ships its own in `Lucebox/DeepSeek-V4.1-Flash-1017-ROCMFP-GGUF`. The engine refuses an encoder from another checkpoint than the target's.

主な特徴

特点摘要待依据来源补齐,不凭模型名称推断能力。

制約と注意点

具体的な制約を判断する出典が不足しています。実際の性能、遅延、データ方針、ライセンス、提供期間を確認してください。

PRICING & COST

出典別料金と費用分析

料金はチャネル・地域・設定によって異なります。各チャネルの料金は個別に表示し、提供元の統一公式料金とは見なしません。

出典・チャネル入力出力課金単位収集日時
確認可能な料金がありません。出典を確認してください。

月額の基本トークン費用を試算

—

試算はキャッシュ割引、段階料金、マルチモーダル追加料金、検索、税金、追加推論出力を含みません。

INDEPENDENT BENCHMARKS

公開評価と実行設定

暂无已采集的可核验公开评分。模型未上榜不代表能力较低。

収録済みの全ランキング →

よくある質問

オープンウェイトなら無料で商用利用できますか?

必ずしもそうではありません。バージョン別のライセンス、商用条件、利用方針を確認してください。自前運用には計算資源と保守費用も必要です。

これはリアルタイムの請求料金ですか?

収集日時付きの料金スナップショットです。最新のバージョン、地域、割引、課金方式は各チャネルで確認してください。

出典と追跡記録

许可证:mit。模型介绍与第三方评测均以原来源的具体版本和方法为准。