gpt-oss-120b VRAM & Memory Requirements
Specific checkpoints with sourced architecture and weight files. MXFP4 safetensors: 61.68 GiB–63.56 GiB at 8,192 tokens, 1 sequence, FP16 KV, vllm.
This result is an estimate, not a measured trace.Not verifiedFile-backed weights
Publisher and provenance
- PublisherOpenAI
- Hugging Face
openai/gpt-oss-120b - Revision
b5c939de8f754692c1647ca79fbf85e8c1e70f8a - Total parameters116,829,156,672
- Active parameters5,100,000,000
- Layers36
- Query heads64
- KV heads8 (sliding-window)
- Head dimension64
- Recorded context capability131,072
- Completenesscomplete · full
Active parameters describe tokens' routed compute, not the full resident model weight footprint. Full-resident weight estimates use total parameters.
Memory layers in this estimate
Artifact bytes, raw KV, engine extras, and total demand are separate. This is not a runtime trace.
- Weight artifact
- 60.77 GiB (65,248,815,744 bytes)
- Raw KV cache
- 0.29 GiB (306,708,480 bytes)
- Engine extras / workspace
- 0.63 GiB–2.51 GiB
- Total estimated demand
- 61.68 GiB–63.56 GiB
Published artifacts
| Quantization | Selected files | Bytes | Engine | Evidence |
|---|---|---|---|---|
| MXFP4 safetensors mxfp4 | 65,248,815,744 (60.77 GiB / 65.25 GB) | vllm, unspecified | file-metadata | |
| INT4 theoretical | No published file | 58,414,578,336 (theoretical) | — | theoretical |
Memory by quantization and context
Concurrent sequences = 1. KV defaults to FP16. The range includes documented runtime extras for the listed engine. These figures use the same core as the interactive calculator.
| Quantization | 2,048 tokens | 8,192 tokens | 32,768 tokens |
|---|---|---|---|
| MXFP4 safetensors | 61.47 GiB–63.34 GiB | 61.68 GiB–63.56 GiB | 62.52 GiB–64.43 GiB |
| INT4 theoretical Theoretical rows are not downloadable artifacts. | 54.60 GiB–70.73 GiB | 54.81 GiB–71.00 GiB | 55.66 GiB–72.04 GiB |
Raw KV contribution
Related pages
Sources
- gpt-oss-120b model card
- config.json — revision b5c939de8f754692c1647ca79fbf85e8c1e70f8a
- safetensors index — metadata.total_size=65248815744
Assumptions that affect this result
- Requested context is within the recorded native window of 131072 tokens.
- Full-resident weight estimates use total parameters, including inactive MoE experts.
- This MoE lists 116829156672 total parameters and 5100000000 active parameters. Active-parameter compute is not used as the weight-residency figure.
- Weight bytes come from the selected weight files only, not from every file in the repository.
- Download file size is not peak GPU memory. Runtime layout, allocator padding, and KV are extra.
- Weight bytes come from 15 published safetensors shards totaling 65248815744 bytes.
- Context budget is the number of cached tokens per sequence (prompt plus reserved generation).
- Concurrent sequences share one model replica; weights are not multiplied by sequence count.
Calculator
Loading catalog snapshot…
Search or pick a catalog checkpoint to see memory demand.
Search or pick a catalog checkpoint to see memory demand.