{"v":1,"id":"model:hf:handy-computer/gigaam-v3-e2e-rnnt-gguf","slug":"model-handy-computer-gigaam-v3-e2e-rnnt-gguf","kind":"model","category":"speech","title":"gigaam-v3-e2e-rnnt-gguf","summary":"GGUF conversions of ai-sage/GigaAM-v3 for use with transcribe.cpp.","source":{"provider":"hf","ref":"handy-computer/gigaam-v3-e2e-rnnt-gguf","url":"https://huggingface.co/handy-computer/gigaam-v3-e2e-rnnt-gguf","rev":"b9b68a835993df09018237a11d2a9b8c1925844d","fetchedAt":"2026-10-02T20:59:51.528Z","etag":"W/\"bc3-AqwpN3BQ6Ztt7m2E9Zh4DbQudMs\""},"author":{"name":"handy-computer","url":"https://huggingface.co/handy-computer"},"license":{"spdx":"mit","raw":"mit","open":true},"metrics":{"downloads":75438,"downloadsWeek":15195,"likes":2,"takenAt":"2026-10-02T20:59:51.528Z"},"tags":["transcribe.cpp","gguf","asr","speech-to-text","gigaam","conformer","rnnt","russian","automatic-speech-recognition","ru"],"pipeline":"automatic-speech-recognition","links":{"github":"ggml-org/whisper.cpp","npm":"nodejs-whisper"},"updatedAt":"2026-09-15T07:05:32.000Z","collectedAt":"2026-10-02T20:59:51.528Z","review":{"numbers":["75,438 downloads on Hugging Face","2 likes","license mit","0.2 GB for gigaam-v3-e2e-rnnt-Q4_K_M.gguf","15,195 npm downloads a week for nodejs-whisper","latest nodejs-whisper@0.3.1"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T20:59:51.528Z","http":200},"description":"# GigaAM-v3: transcribe.cpp GGUF\n\nGGUF conversions of [ai-sage/GigaAM-v3](https://huggingface.co/ai-sage/GigaAM-v3) for use\nwith [transcribe.cpp](https://github.com/handy-computer/transcribe.cpp).\n\nPorted from upstream commit\n[c7f128b](https://huggingface.co/ai-sage/GigaAM-v3/commit/c7f128b),\npinned 2026-05-12.\nValidated against the gigaam author package reference at transcribe.cpp commit\n[42b96d9](https://github.com/handy-computer/transcribe.cpp/tree/42b96d9)\non 2026-05-12.\n\nOffline Russian speech-to-text with greedy RNN-T decoding. Same 16-layer Conformer encoder as the e2e variant, fine-tuned to emit lowercased Russian with no punctuation; 33-entry character vocabulary.\n\n## Downloads\n\n| Quantization | Download | Size | WER (FLEURS ru) |\n| --- | --- | ---: | ---: |\n| F32 | [gigaam-v3-rnnt-F32.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-F32.gguf) | 888 MB | 8.08% |\n| F16 | [gigaam-v3-rnnt-F16.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-F16.gguf) | 451 MB | 8.08% |\n| Q8_0 | [gigaam-v3-rnnt-Q8_0.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-Q8_0.gguf) | 273 MB | 8.08% |\n| Q6_K | [gigaam-v3-rnnt-Q6_K.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-Q6_K.gguf) | 227 MB | 8.07% |\n| Q5_K_M | [gigaam-v3-rnnt-Q5_K_M.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-Q5_K_M.gguf) | 206 MB | 8.12% |\n| Q4_K_M | [gigaam-v3-rnnt-Q4_K_M.gguf](https://huggingface.co/handy-computer/gigaam-v3-rnnt-gguf/resolve/main/gigaam-v3-rnnt-Q4_K_M.gguf) | 183 MB | 8.12% |\n\nWER on the full FLEURS ru split (775 utterances), batch size 1, timestamps none. Figures without a commit were published before provenance was recorded.\n\nGreedy decoding, no external LM. F32 reference baseline: 8.08%. Upstream `gig…\n\nSource: https://huggingface.co/handy-computer/gigaam-v3-e2e-rnnt-gguf","install":{"kind":"model","hfId":"handy-computer/gigaam-v3-e2e-rnnt-gguf","gated":false,"format":"gguf","files":[{"name":"gigaam-v3-e2e-rnnt-F16.gguf","size":452381408,"quant":"F16","sha256":"87b12fa3523939d3c47c4b427f587e415dda41dcafa9e4ba17f2a1cab2f0d5ad"},{"name":"gigaam-v3-e2e-rnnt-F32.gguf","size":890138592,"quant":"F32","sha256":"0de6fa626b8a3429dc80e4b34c4b03c0e85b02e2eca4c654efbeb27a8a0a4c5b"},{"name":"gigaam-v3-e2e-rnnt-Q4_K_M.gguf","size":183948704,"quant":"Q4_K_M","sha256":"7d69952fb431a8d7800ed9910dc61fea37d8406bfe96d10bf24c8bd4b7c68623"},{"name":"gigaam-v3-e2e-rnnt-Q5_K_M.gguf","size":206392736,"quant":"Q5_K_M","sha256":"0f5969b05f7e196e8922e243c64963132025a5ae7ec2f83d028ae120d201594f"},{"name":"gigaam-v3-e2e-rnnt-Q6_K.gguf","size":227953952,"quant":"Q6_K","sha256":"1ef0f771d9ab0b0675c45ce67a3e5b79ac7497fe35ebb08a51704c4ee423edce"},{"name":"gigaam-v3-e2e-rnnt-Q8_0.gguf","size":273724832,"quant":"Q8_0","sha256":"78d63b47723b7f8d78c6113a6ef983b5a86e2a86f6c273e1f5cb6967b1c4467a"}],"totalBytes":2234540224,"suggestedFile":"gigaam-v3-e2e-rnnt-Q4_K_M.gguf","requirements":{"ramGb":1,"diskBytes":183948704,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["whisper.cpp"],"command":"lsh models install hf:handy-computer/gigaam-v3-e2e-rnnt-gguf"}}