GGUF, Safetensors & NPY Viewer (NumPy Viewer)

For people who run or fine-tune local models, Tensor Viewer opens .gguf and .safetensors files in VS Code, works as a NumPy viewer for .npy and .npz, and shows every tensor's name, dtype and shape, plus stats and a heatmap, without Python.

A real VS Code 1.142 window running this extension on test/fixtures/tiny-llama-q8.gguf, a 3.35 MB llama-style GGUF written by gguf-py from scripts/make-fixtures.py.

• It opens big checkpoints in well under a second and shows the real numbers. Only the header is read at first, and a click on a tensor reads that tensor's bytes by offset. Our test opens a 10.7 GB safetensors file in under 1 s. The Values table pages through any tensor, one 2-D slice at a time, and Save writes a tensor or a slice to .npy or CSV.
• It reads the formats local LLM users actually have. GGUF metadata (architecture, context length, file type, tokenizer, chat template) sits next to the quant type of each tensor. Q8_0, Q4_0, Q4_1, Q5_0, Q5_1, Q4_K, Q5_K and Q6_K weights are decoded for stats and heatmaps.
• It's safe to open files you downloaded. Nothing in the file is ever run. Pickled object arrays in .npy and .npz are listed and refused, and the viewer makes no network calls at all.

Quick start

1. Install Tensor Viewer.
2. Click any .gguf, .safetensors, .npy or .npz file in the Explorer. Then click a tensor in the table.

Recorded in the same window. The white arrow is drawn in to show where each click lands.

Run "Tensor Viewer: Get Started" from the Command Palette for a short tour.

What you see

The top bar shows the total parameter count, the number of tensors, the file size and how long the header took. GGUF files add a summary row with the architecture, model name, context length, embedding size, layers, heads, file type, tokenizer size and bits per weight.

The tensor table lists name, type, shape, parameter count, size in bytes and byte offset. Type in the filter box to match names (attn_q, blk.3.) or an exact type (Q4_K, float16). Click a header to sort. Up and down arrows move through the list. Tables with thousands of tensors show 500 rows at a time.

Click a tensor and the right panel shows min, max, mean, std, NaN, +Inf, -Inf and zero counts. 2-D tensors get a heatmap. 1-D tensors get a line plot with a min and max band, and an HxWx3 uint8 array shows as a picture. A histogram comes last. Hover over any of them to read exact values. "Copy shape" puts (4096, 11008) on the clipboard, and "Copy torch.Size" gives torch.Size([4096, 11008]).

Values table

Under the heatmap, the Values table shows the numbers themselves, 32 rows at a time. Use the arrow buttons to page through rows and columns, or type a row or column number and press Enter. The number format can be Auto, Exact, 2, 4 or 8 decimals, or Scientific. Hover over a cell to see its index and its exact value. "Copy values" puts the page on the clipboard as tab-separated text, which pastes into a spreadsheet.

GGUF quant types such as Q4_K and Q8_0 show their values after dequantizing, as llama.cpp computes them. float16, bfloat16 and FP8 values show as their exact float32 value. Complex numbers show as 3+4j.

For tensors with 3 or more dimensions, the Slice field picks one 2-D slice, for example [5, :, :] for the sixth image in a batch. The table shows that slice. Tick "Use only this slice for the stats, the heatmap and Save" and the stats, the heatmap and the Save buttons use it too.

Save as .npy or CSV

"Save as .npy..." writes the tensor (or the slice) to a NumPy file that np.load reads. Types NumPy has, such as float32, int64 or complex64, are copied byte for byte. bfloat16, FP8 and the GGUF quant types are written as float32. "Save as CSV..." writes one row per line with every digit kept. 3-D and bigger tensors become rows of their last dimension. You pick the file in a save dialog, the file is written in chunks with a progress bar you can cancel, and Tensor Viewer never writes over the file it is showing.

The heatmap uses blue for negative values and red for positive ones, with gray at zero. Data that's never negative uses a single blue ramp instead. Colors are clipped at the 99.5th percentile so a few outlier weights don't flatten everything else, and the caption says when that happened.

Supported files

Other GGUF types (Q2_K, Q3_K, IQ types, TQ, MXFP4) are listed with the right size and offset, but have no stats yet. Text, structured and object arrays in NumPy files are listed without stats.

The GGUF shapes are shown in row-major order, the way PyTorch prints them. GGUF itself stores the dimensions fastest-first, so [4096, 32000] in llama.cpp logs is [32000, 4096] here.

How big files stay fast

Tensor Viewer never reads a whole file. The .npy and safetensors headers are a few hundred bytes to a few MB, and GGUF metadata is read in 2 MB steps until the tensor list ends. For one tensor it reads at most tensorViewer.sampleValues values (2,000,000 by default). When a tensor is bigger, it reads up to 512 runs of whole rows spread evenly from the first row to the last, so the heatmap covers the full tensor. The panel always says whether the stats are exact or from a sample, and how many values went in.

Deflated .npz entries can't be read at an offset. For those, the header comes from inflating just the start of the entry, and big arrays are sampled from their first rows.

The Values table reads only the cells on the page, so a page of a 10 GB tensor costs a few KB.

Privacy and safety

There's no telemetry and no network access. The webview runs under a strict Content-Security-Policy with no remote scripts, and every name and metadata string is drawn as text, never as HTML. Python pickle is never loaded, so .pt, .pth, .pkl and object arrays inside .npy files stay closed. Tensor Viewer never changes the file it shows. Save as .npy and Save as CSV write only to a new file that you pick.

Limits

GGUF files split into shards show the tensors of the open shard only, with a note. Heatmaps show at most 256 by 256 cells, each the mean of its block of values. Tensors with three or more dimensions are drawn with the leading dimensions flattened into rows, unless you tick "Use only this slice". Slices of Fortran-order arrays show in the Values table, but the stats and heatmap cover the whole array. CSV export stops at 20 million values; use .npy for bigger tensors. Compressed .npz entries can only be read from the start, so values past their first 64 MB are not shown, and entries over 512 MB can't be saved. 64-bit integers above 2^53 lose their last digits in the table and in CSV. The .npy export copies them exactly.

License

MIT. Bundled libraries are listed in THIRD-PARTY-NOTICES.md (@huggingface/gguf, @huggingface/tasks and fflate, all MIT).

Support

Zovo extensions have no ads and no tracking, and sponsors pay for the time that goes into fixes and new viewers. You can sponsor the work on GitHub Sponsors.

More viewers by Zovo

• HDF5 & NetCDF Explorer (H5 Viewer) browses groups, attributes and values in HDF5 and NetCDF files and plots slices.
• CSV Viewer, Excel Viewer (XLSX) & Parquet Viewer opens CSV, TSV, Excel and Parquet files as a sortable, filterable table.
• PDF Viewer & PDF Reader with Dark Mode reads PDFs with a real dark mode, search, an outline and thumbnails.
• DOCX Viewer & Word Viewer with Dark Mode previews Word and ODT files with dark mode, an outline and search.
• PPTX Viewer & PowerPoint Viewer shows PowerPoint slides, thumbnails and speaker notes.
• Large JSON Viewer & JSON Tree shows large JSON files as a searchable tree, copies JSONPath and turns arrays of objects into a table.
• JSONL Viewer & NDJSON Viewer (JSONL Table) opens JSONL and NDJSON files as a table, as records or as a chat view.
• SQLite Viewer & DB Viewer (Read-Only) browses SQLite tables read-only and runs SELECT queries.
• MP4 Video Player & Video Preview with Sound plays MP4, MOV, MKV and WebM files with their sound.
• DXF Viewer & CAD Viewer for 2D Drawings opens ASCII and binary DXF drawings with layers, measuring and PNG or SVG export.
• AGENTS.md & CLAUDE.md Linter (Context Drift) finds stale commands, paths and versions in AGENTS.md and CLAUDE.md files.
• Agent Skills & Claude Skills (Skill Shelf) lists, checks and toggles the Agent Skills that Claude Code, Codex and Cursor load.
• Claude Code History & Codex History Viewer browses and searches your local Claude Code and Codex sessions.
• Claude Code Usage & Codex Usage Cost Meter shows Claude Code and Codex tokens and estimated cost from your local session files.

Coming soon to the Marketplace, with the code on GitHub now.

• PKL Viewer & Pickle Viewer opens Python pickle files without Python and never runs their code.
• Audio Preview & Loudness Meter plays audio files with a waveform, a spectrogram and LUFS loudness.
• Graphviz Preview & DOT Viewer shows a live preview of DOT files and exports SVG or PNG.
• TIFF Viewer (16-bit, Float, GeoTIFF) opens 16-bit, float, multi-page and GeoTIFF images.
• Avro Viewer, Arrow Viewer & Feather Viewer opens Avro, Arrow and Feather files as a paged table.
• EML Viewer & MBOX Email Viewer reads EML, EMLX and MBOX email files with remote images blocked.
• 3D Model Viewer (GLB, glTF, STL, OBJ, PLY) views GLB, glTF, STL, OBJ, PLY and 3MF models.
• DICOM Viewer & NIfTI Viewer for CT and MRI opens CT and MRI scans, DICOM folders and NIfTI volumes with window presets and three views.
• ZIP Viewer & Archive Explorer (JAR, TAR, GZ) browses ZIP, JAR, VSIX, TAR and GZ files, opens entries in tabs and extracts them safely.
• GCode Viewer & 3D Print Preview shows GCode toolpaths in 3D layer by layer, with print time and filament use.
• EPUB Reader & EPUB Viewer reads EPUB books with a table of contents, search and light, sepia or dark themes.
• HAR Viewer & HTTP Archive Viewer opens HAR files from browser DevTools with a timing waterfall, filters and a redacted copy for sharing.
• Font Viewer & Font Preview (TTF, OTF, WOFF) opens TTF, OTF, WOFF, WOFF2 and TTC fonts with a glyph grid, a type tester with variable axes and the license fields.
• File Viewers Pack installs the PDF, table, Word, PowerPoint, HDF5 and tensor viewers in one step.