This is an exllamav3 quantization of orcarouter’s GLM-5.3-Flash-Uncensored-FP8. Quantized targeting 2.5 bpw with the -hq option to get 2.51. 2.51bpwh6 98.123 GiB (This model) 3.05bpwh6 117.529 GiB 4.05bpw_h6 153.810 GiB > The abliterated (refusal-removed) build of zai-org/GLM-5.3-Flash > — a 320B / 18B-active Mixture-of-Experts model with hybrid linear + sparse attention, > 4-wide Manifold-Constrained Hyper-Connections, a native vision + video tower, an MTP > speculative head, and a 1M-token context. > > The refusal direction is baked directly into the official block-FP8 shards — same format, same > shard layout, same model.safetensors.index.json. Every tensor matches the base checkpoint in > name, dtype and shape, so this is a drop-in replacement for zai-org/GLM-5.3-Flash in any > stack that already serves it. > > On the -FP8 name: block-FP8 is the format Z.ai ships GLM-5.3-Flash in — the suffix names the > checkpoint’s own precision, not a quantization step we applied. There is no upstream BF16 release to > derive from, so this is the full-precision source as published. OrcaCode Review turns every model listed on OrcaRouter into a production code-review agent: — reviews every PR…
Модальности:
Генерация текста
Области применения:
Логика и рассуждение Диалог / чат Вызов функций (Tool use)
Задача: Генерация текста
Автор: MikeRoz
Теги: exllamav3, glm5_next, abliterated, glm, glm5, glm-5.3-flash, uncensored, ai-red-team
Лайков: 3 | Загрузок: 525
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
