nvidia · observed 19 Oct 2026
Public-source support evidence
nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB
No description recorded in this snapshot.
View source model ↗SUPPORT EVIDENCE
The exact config architecture appears in llama.cpp MMPROJ_MODEL_MAP only. Projector conversion is listed; full text model conversion and runtime support are unverified.
Checked 2026-10-19 02:18 UTC
Evidence source ↗In release. Fresh Ollama curated library tags page exposes this exact local model variant; catalog availability only, not hardware-tested support.
Checked 2026-10-19 10:01 UTC
Evidence source ↗Fresh public repository README contains the exact model id or full title and the repository contains GGUF weight files. The repository omits base_model metadata, so this supports community upload availability only; no exact linked catalog release was verified.
Checked 2026-10-19 10:01 UTC
Evidence source ↗Targeted public search of mlx-community yielded no exact base_model match with MLX file metadata; absence does not mean unsupported.
Checked 2026-10-19 10:01 UTC
Checked source ↗The model config architecture is listed in this runtime registry. This is family/architecture evidence only; exact checkpoint loading, model-specific configuration, hardware, and generation were not tested.
Checked 2026-10-19 02:18 UTC
Evidence source ↗MLX community uploads may target mlx-vlm; an upload observation does not establish mlx-lm compatibility.
MEMORY ESTIMATES
Which file should I grab?
No observed complete artifact currently qualifies.Unknown: no observed complete artifact fits this capacity with documented headroom.
Which file should I grab?
No observed complete artifact currently qualifies.Unknown: no observed complete artifact fits this capacity with documented headroom.
Which file should I grab?
No observed complete artifact currently qualifies.Unknown: no observed complete artifact fits this capacity with documented headroom.
Which file should I grab?
No observed complete artifact currently qualifies.Unknown: no observed complete artifact fits this capacity with documented headroom.
Observed artifact size plus 20% headroom and 1.5 GB runtime allowance. Fit is an estimate; it does not establish runtime compatibility, speed, context length, or hardware performance. Text-weights-only estimates do not include a projector or total vision memory where unknown.
OBSERVED FILES
Sizes and file groups come from the listed uploader repositories. A size is marked as an estimated fit only for a complete observed GGUF quant group.
No observed quant uploads in this snapshot.
CURATED TRACKING
No curated issues verified for this snapshot. An empty list does not mean there are no bugs.
HARDWARE NOTES
No hardware benchmark measurements are available in this snapshot.