Model field guide
granite-4.1-8b
42 of 44 hardware configurations have a supported memory fit at one or more recorded quantizations. Runtime compatibility still needs verification.
Total parameters8.79B
Active parameters8.79B
Recorded context limit131,072
Architecturedense
Verified 2026-10-22 · license metadata: apache-2.0 · commercial use: yes (review the terms).
Actual files, not a size shortcut.
| Quantization | Total file size | Source |
|---|---|---|
| Q4_K_M | 4.981 GiB5,347,914,400 bytes; 1 file(s)Sum of exact file byte sizes in the repository listing at the recorded commit; every declared shard is present. Auxiliary vision projector and speculative-decoding files are excluded. Labels follow publisher filenames; mixed-precision expert tensors may remain in a differently labeled file, so the estimator uses measured bytes rather than assumed bits per weight. | Pinned file recordVerified 2026-10-22 |
| Q5_K_M | 5.824 GiB6,253,884,064 bytes; 1 file(s)Sum of exact file byte sizes in the repository listing at the recorded commit; every declared shard is present. Auxiliary vision projector and speculative-decoding files are excluded. Labels follow publisher filenames; mixed-precision expert tensors may remain in a differently labeled file, so the estimator uses measured bytes rather than assumed bits per weight. | Pinned file recordVerified 2026-10-22 |
| Q8_0 | 8.704 GiB9,345,610,400 bytes; 1 file(s)Sum of exact file byte sizes in the repository listing at the recorded commit; every declared shard is present. Auxiliary vision projector and speculative-decoding files are excluded. Labels follow publisher filenames; mixed-precision expert tensors may remain in a differently labeled file, so the estimator uses measured bytes rather than assumed bits per weight. | Pinned file recordVerified 2026-10-22 |
The hardware picture.
Q4_K_M: hardware comparison
| Hardware | Usable memory | Required memory | Fit estimate | Estimated generation |
|---|---|---|---|---|
| Mac mini M6 (12-core GPU) 16GB | 12.0 GiB | 6.61 GiB | fits | 11.4–22.9 estimated tokens/s |
| Mac mini M6 (12-core GPU) 24GB | 18.0 GiB | 6.61 GiB | fits | 12.7–25.4 estimated tokens/s |
| Mac mini M6 (12-core GPU) 32GB | 24.0 GiB | 6.61 GiB | fits | 12.7–25.4 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 24GB | 18.0 GiB | 6.61 GiB | fits | 23.0–45.9 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 48GB | 36.0 GiB | 6.61 GiB | fits | 23.0–45.9 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 64GB | 48.0 GiB | 6.61 GiB | fits | 23.0–45.9 estimated tokens/s |
| Mac Studio M5 Max (32-core GPU) 36GB | 27.0 GiB | 6.61 GiB | fits | 34.4–68.8 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 64GB | 48.0 GiB | 6.61 GiB | fits | 45.9–91.8 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 128GB | 96.0 GiB | 6.61 GiB | fits | 45.9–91.8 estimated tokens/s |
| Mac Studio M5 Ultra (30-core CPU / 64-core GPU) 96GB | 72.0 GiB | 6.61 GiB | fits | 89.8–179.5 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 256GB | 192.0 GiB | 6.61 GiB | fits | 89.8–179.5 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 512GB | 384.0 GiB | 6.61 GiB | fits | 89.8–179.5 estimated tokens/s |
| MacBook Pro 14-inch M5 (10-core GPU) 24GB | 18.0 GiB | 6.61 GiB | fits | 11.4–22.9 estimated tokens/s |
| MacBook Pro 14-inch M5 Pro (20-core GPU) 48GB | 36.0 GiB | 6.61 GiB | fits | 23.0–45.9 estimated tokens/s |
| MacBook Pro 16-inch M5 Max (40-core GPU) 128GB | 96.0 GiB | 6.61 GiB | fits | 45.9–91.8 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 16GB | 12.0 GiB | 6.61 GiB | fits | 9.0–18.0 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 24GB | 18.0 GiB | 6.61 GiB | fits | 9.0–18.0 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 32GB | 24.0 GiB | 6.61 GiB | fits | 9.0–18.0 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 24GB | 18.0 GiB | 6.61 GiB | fits | 20.4–40.8 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 48GB | 36.0 GiB | 6.61 GiB | fits | 20.4–40.8 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 64GB | 48.0 GiB | 6.61 GiB | fits | 20.4–40.8 estimated tokens/s |
| Mac Studio (2025) M4 Max (32-core GPU) 36GB | 27.0 GiB | 6.61 GiB | fits | 30.7–61.3 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 64GB | 48.0 GiB | 6.61 GiB | fits | 40.8–81.7 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 128GB | 96.0 GiB | 6.61 GiB | fits | 40.8–81.7 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (28-core CPU / 60-core GPU) 96GB | 72.0 GiB | 6.61 GiB | fits | 61.3–122.5 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (32-core CPU / 80-core GPU) 256GB | 192.0 GiB | 6.61 GiB | fits | 61.3–122.5 estimated tokens/s |
| NVIDIA GeForce RTX 3060 12GB | 10.8 GiB | 6.61 GiB | fits | 26.9–53.9 estimated tokens/s |
| NVIDIA GeForce RTX 3090 24GB | 21.6 GiB | 6.61 GiB | fits | 70.0–140.0 estimated tokens/s |
| NVIDIA GeForce RTX 4060 Ti 16GB | 14.4 GiB | 6.61 GiB | fits | 21.5–43.1 estimated tokens/s |
| NVIDIA GeForce RTX 4070 12GB GDDR6X | 10.8 GiB | 6.61 GiB | fits | 37.7–75.4 estimated tokens/s |
| NVIDIA GeForce RTX 4080 16GB | 14.4 GiB | 6.61 GiB | fits | 53.6–107.2 estimated tokens/s |
| NVIDIA GeForce RTX 4090 24GB | 21.6 GiB | 6.61 GiB | fits | 75.4–150.8 estimated tokens/s |
| NVIDIA GeForce RTX 5060 Ti 16GB | 14.4 GiB | 6.61 GiB | fits | 33.5–67.0 estimated tokens/s |
| NVIDIA GeForce RTX 5070 12GB | 10.8 GiB | 6.61 GiB | fits | 50.3–100.5 estimated tokens/s |
| NVIDIA GeForce RTX 5070 Ti 16GB | 14.4 GiB | 6.61 GiB | fits | 67.0–134.0 estimated tokens/s |
| NVIDIA GeForce RTX 5080 16GB | 14.4 GiB | 6.61 GiB | fits | 71.8–143.6 estimated tokens/s |
| NVIDIA GeForce RTX 5090 32GB | 28.8 GiB | 6.61 GiB | fits | 134.0–268.1 estimated tokens/s |
| NVIDIA RTX PRO 6000 Blackwell Workstation Edition 96GB | 86.4 GiB | 6.61 GiB | fits | 134.0–268.1 estimated tokens/s |
| NVIDIA RTX 6000 Ada Generation 48GB | 43.2 GiB | 6.61 GiB | fits | 71.8–143.6 estimated tokens/s |
| Intel Arc B580 12GB | 10.8 GiB | 6.61 GiB | fits | 34.1–68.2 estimated tokens/s |
| AMD Ryzen AI Halo, Ryzen AI Max+ 395, 128GB | 89.6 GiB | 6.61 GiB | fits | 19.1–38.3 estimated tokens/s |
| NVIDIA DGX Spark 128GB / 4TB | 96.0 GiB | 6.61 GiB | fits | 20.4–40.8 estimated tokens/s |
| 2 × GeForce RTX 3090 24GB (conditional 48GB aggregate) | 40.8 GiB | 6.61 GiB | Needs verification | Estimated speed unavailable |
| 2 × GeForce RTX 5090 32GB (conditional 64GB aggregate) | 54.4 GiB | 6.61 GiB | Needs verification | Estimated speed unavailable |
Q5_K_M: hardware comparison
| Hardware | Usable memory | Required memory | Fit estimate | Estimated generation |
|---|---|---|---|---|
| Mac mini M6 (12-core GPU) 16GB | 12.0 GiB | 7.45 GiB | fits | 9.8–19.6 estimated tokens/s |
| Mac mini M6 (12-core GPU) 24GB | 18.0 GiB | 7.45 GiB | fits | 10.9–21.7 estimated tokens/s |
| Mac mini M6 (12-core GPU) 32GB | 24.0 GiB | 7.45 GiB | fits | 10.9–21.7 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 24GB | 18.0 GiB | 7.45 GiB | fits | 19.6–39.3 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 48GB | 36.0 GiB | 7.45 GiB | fits | 19.6–39.3 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 64GB | 48.0 GiB | 7.45 GiB | fits | 19.6–39.3 estimated tokens/s |
| Mac Studio M5 Max (32-core GPU) 36GB | 27.0 GiB | 7.45 GiB | fits | 29.4–58.8 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 64GB | 48.0 GiB | 7.45 GiB | fits | 39.3–78.5 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 128GB | 96.0 GiB | 7.45 GiB | fits | 39.3–78.5 estimated tokens/s |
| Mac Studio M5 Ultra (30-core CPU / 64-core GPU) 96GB | 72.0 GiB | 7.45 GiB | fits | 76.8–153.5 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 256GB | 192.0 GiB | 7.45 GiB | fits | 76.8–153.5 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 512GB | 384.0 GiB | 7.45 GiB | fits | 76.8–153.5 estimated tokens/s |
| MacBook Pro 14-inch M5 (10-core GPU) 24GB | 18.0 GiB | 7.45 GiB | fits | 9.8–19.6 estimated tokens/s |
| MacBook Pro 14-inch M5 Pro (20-core GPU) 48GB | 36.0 GiB | 7.45 GiB | fits | 19.6–39.3 estimated tokens/s |
| MacBook Pro 16-inch M5 Max (40-core GPU) 128GB | 96.0 GiB | 7.45 GiB | fits | 39.3–78.5 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 16GB | 12.0 GiB | 7.45 GiB | fits | 7.7–15.4 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 24GB | 18.0 GiB | 7.45 GiB | fits | 7.7–15.4 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 32GB | 24.0 GiB | 7.45 GiB | fits | 7.7–15.4 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 24GB | 18.0 GiB | 7.45 GiB | fits | 17.5–34.9 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 48GB | 36.0 GiB | 7.45 GiB | fits | 17.5–34.9 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 64GB | 48.0 GiB | 7.45 GiB | fits | 17.5–34.9 estimated tokens/s |
| Mac Studio (2025) M4 Max (32-core GPU) 36GB | 27.0 GiB | 7.45 GiB | fits | 26.2–52.4 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 64GB | 48.0 GiB | 7.45 GiB | fits | 34.9–69.8 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 128GB | 96.0 GiB | 7.45 GiB | fits | 34.9–69.8 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (28-core CPU / 60-core GPU) 96GB | 72.0 GiB | 7.45 GiB | fits | 52.4–104.8 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (32-core CPU / 80-core GPU) 256GB | 192.0 GiB | 7.45 GiB | fits | 52.4–104.8 estimated tokens/s |
| NVIDIA GeForce RTX 3060 12GB | 10.8 GiB | 7.45 GiB | fits | 23.0–46.1 estimated tokens/s |
| NVIDIA GeForce RTX 3090 24GB | 21.6 GiB | 7.45 GiB | fits | 59.9–119.7 estimated tokens/s |
| NVIDIA GeForce RTX 4060 Ti 16GB | 14.4 GiB | 7.45 GiB | fits | 18.4–36.8 estimated tokens/s |
| NVIDIA GeForce RTX 4070 12GB GDDR6X | 10.8 GiB | 7.45 GiB | fits | 32.2–64.5 estimated tokens/s |
| NVIDIA GeForce RTX 4080 16GB | 14.4 GiB | 7.45 GiB | fits | 45.8–91.7 estimated tokens/s |
| NVIDIA GeForce RTX 4090 24GB | 21.6 GiB | 7.45 GiB | fits | 64.5–128.9 estimated tokens/s |
| NVIDIA GeForce RTX 5060 Ti 16GB | 14.4 GiB | 7.45 GiB | fits | 28.7–57.3 estimated tokens/s |
| NVIDIA GeForce RTX 5070 12GB | 10.8 GiB | 7.45 GiB | fits | 43.0–86.0 estimated tokens/s |
| NVIDIA GeForce RTX 5070 Ti 16GB | 14.4 GiB | 7.45 GiB | fits | 57.3–114.6 estimated tokens/s |
| NVIDIA GeForce RTX 5080 16GB | 14.4 GiB | 7.45 GiB | fits | 61.4–122.8 estimated tokens/s |
| NVIDIA GeForce RTX 5090 32GB | 28.8 GiB | 7.45 GiB | fits | 114.6–229.2 estimated tokens/s |
| NVIDIA RTX PRO 6000 Blackwell Workstation Edition 96GB | 86.4 GiB | 7.45 GiB | fits | 114.6–229.2 estimated tokens/s |
| NVIDIA RTX 6000 Ada Generation 48GB | 43.2 GiB | 7.45 GiB | fits | 61.4–122.8 estimated tokens/s |
| Intel Arc B580 12GB | 10.8 GiB | 7.45 GiB | fits | 29.2–58.3 estimated tokens/s |
| AMD Ryzen AI Halo, Ryzen AI Max+ 395, 128GB | 89.6 GiB | 7.45 GiB | fits | 16.4–32.7 estimated tokens/s |
| NVIDIA DGX Spark 128GB / 4TB | 96.0 GiB | 7.45 GiB | fits | 17.5–34.9 estimated tokens/s |
| 2 × GeForce RTX 3090 24GB (conditional 48GB aggregate) | 40.8 GiB | 7.45 GiB | Needs verification | Estimated speed unavailable |
| 2 × GeForce RTX 5090 32GB (conditional 64GB aggregate) | 54.4 GiB | 7.45 GiB | Needs verification | Estimated speed unavailable |
Q8_0: hardware comparison
| Hardware | Usable memory | Required memory | Fit estimate | Estimated generation |
|---|---|---|---|---|
| Mac mini M6 (12-core GPU) 16GB | 12.0 GiB | 10.33 GiB | fits | 6.5–13.1 estimated tokens/s |
| Mac mini M6 (12-core GPU) 24GB | 18.0 GiB | 10.33 GiB | fits | 7.3–14.6 estimated tokens/s |
| Mac mini M6 (12-core GPU) 32GB | 24.0 GiB | 10.33 GiB | fits | 7.3–14.6 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 24GB | 18.0 GiB | 10.33 GiB | fits | 13.1–26.3 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 48GB | 36.0 GiB | 10.33 GiB | fits | 13.1–26.3 estimated tokens/s |
| Mac mini M5 Pro (15-core CPU / 16-core GPU) 64GB | 48.0 GiB | 10.33 GiB | fits | 13.1–26.3 estimated tokens/s |
| Mac Studio M5 Max (32-core GPU) 36GB | 27.0 GiB | 10.33 GiB | fits | 19.7–39.4 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 64GB | 48.0 GiB | 10.33 GiB | fits | 26.3–52.6 estimated tokens/s |
| Mac Studio M5 Max (40-core GPU) 128GB | 96.0 GiB | 10.33 GiB | fits | 26.3–52.6 estimated tokens/s |
| Mac Studio M5 Ultra (30-core CPU / 64-core GPU) 96GB | 72.0 GiB | 10.33 GiB | fits | 51.4–102.7 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 256GB | 192.0 GiB | 10.33 GiB | fits | 51.4–102.7 estimated tokens/s |
| Mac Studio M5 Ultra (36-core CPU / 80-core GPU) 512GB | 384.0 GiB | 10.33 GiB | fits | 51.4–102.7 estimated tokens/s |
| MacBook Pro 14-inch M5 (10-core GPU) 24GB | 18.0 GiB | 10.33 GiB | fits | 6.5–13.1 estimated tokens/s |
| MacBook Pro 14-inch M5 Pro (20-core GPU) 48GB | 36.0 GiB | 10.33 GiB | fits | 13.1–26.3 estimated tokens/s |
| MacBook Pro 16-inch M5 Max (40-core GPU) 128GB | 96.0 GiB | 10.33 GiB | fits | 26.3–52.6 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 16GB | 12.0 GiB | 10.33 GiB | fits | 5.1–10.3 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 24GB | 18.0 GiB | 10.33 GiB | fits | 5.1–10.3 estimated tokens/s |
| Mac mini (2024) M4 (10-core GPU) 32GB | 24.0 GiB | 10.33 GiB | fits | 5.1–10.3 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 24GB | 18.0 GiB | 10.33 GiB | fits | 11.7–23.4 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 48GB | 36.0 GiB | 10.33 GiB | fits | 11.7–23.4 estimated tokens/s |
| Mac mini (2024) M4 Pro (16-core GPU) 64GB | 48.0 GiB | 10.33 GiB | fits | 11.7–23.4 estimated tokens/s |
| Mac Studio (2025) M4 Max (32-core GPU) 36GB | 27.0 GiB | 10.33 GiB | fits | 17.5–35.1 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 64GB | 48.0 GiB | 10.33 GiB | fits | 23.4–46.7 estimated tokens/s |
| Mac Studio (2025) M4 Max (40-core GPU) 128GB | 96.0 GiB | 10.33 GiB | fits | 23.4–46.7 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (28-core CPU / 60-core GPU) 96GB | 72.0 GiB | 10.33 GiB | fits | 35.1–70.1 estimated tokens/s |
| Mac Studio (2025) M3 Ultra (32-core CPU / 80-core GPU) 256GB | 192.0 GiB | 10.33 GiB | fits | 35.1–70.1 estimated tokens/s |
| NVIDIA GeForce RTX 3060 12GB | 10.8 GiB | 10.33 GiB | borderline | 15.4–30.8 estimated tokens/s |
| NVIDIA GeForce RTX 3090 24GB | 21.6 GiB | 10.33 GiB | fits | 40.1–80.1 estimated tokens/s |
| NVIDIA GeForce RTX 4060 Ti 16GB | 14.4 GiB | 10.33 GiB | fits | 12.3–24.7 estimated tokens/s |
| NVIDIA GeForce RTX 4070 12GB GDDR6X | 10.8 GiB | 10.33 GiB | borderline | 21.6–43.1 estimated tokens/s |
| NVIDIA GeForce RTX 4080 16GB | 14.4 GiB | 10.33 GiB | fits | 30.7–61.4 estimated tokens/s |
| NVIDIA GeForce RTX 4090 24GB | 21.6 GiB | 10.33 GiB | fits | 43.1–86.3 estimated tokens/s |
| NVIDIA GeForce RTX 5060 Ti 16GB | 14.4 GiB | 10.33 GiB | fits | 19.2–38.3 estimated tokens/s |
| NVIDIA GeForce RTX 5070 12GB | 10.8 GiB | 10.33 GiB | borderline | 28.8–57.5 estimated tokens/s |
| NVIDIA GeForce RTX 5070 Ti 16GB | 14.4 GiB | 10.33 GiB | fits | 38.3–76.7 estimated tokens/s |
| NVIDIA GeForce RTX 5080 16GB | 14.4 GiB | 10.33 GiB | fits | 41.1–82.2 estimated tokens/s |
| NVIDIA GeForce RTX 5090 32GB | 28.8 GiB | 10.33 GiB | fits | 76.7–153.4 estimated tokens/s |
| NVIDIA RTX PRO 6000 Blackwell Workstation Edition 96GB | 86.4 GiB | 10.33 GiB | fits | 76.7–153.4 estimated tokens/s |
| NVIDIA RTX 6000 Ada Generation 48GB | 43.2 GiB | 10.33 GiB | fits | 41.1–82.2 estimated tokens/s |
| Intel Arc B580 12GB | 10.8 GiB | 10.33 GiB | borderline | 19.5–39.0 estimated tokens/s |
| AMD Ryzen AI Halo, Ryzen AI Max+ 395, 128GB | 89.6 GiB | 10.33 GiB | fits | 11.0–21.9 estimated tokens/s |
| NVIDIA DGX Spark 128GB / 4TB | 96.0 GiB | 10.33 GiB | fits | 11.7–23.4 estimated tokens/s |
| 2 × GeForce RTX 3090 24GB (conditional 48GB aggregate) | 40.8 GiB | 10.33 GiB | Needs verification | Estimated speed unavailable |
| 2 × GeForce RTX 5090 32GB (conditional 64GB aggregate) | 54.4 GiB | 10.33 GiB | Needs verification | Estimated speed unavailable |
A “fits” result checks estimated memory only. “Needs verification” preserves missing cache, vision, or placement information. All speed figures are estimated from the published bandwidth formula.
Field-level sources and verification dates
- id: Official source · verified 2026-10-22. Lowercase URL slug derived from the upstream repository ID; baseModel retains the exact Hugging Face identifier.
- name: Official source · verified 2026-10-22
- baseModel: Official source · verified 2026-10-22
- downloads: Official source · verified 2026-10-22
- tasks: Official source · verified 2026-10-22. Chat denotes a text-generating model, including base models that may require a chat template or instruction tuning. Code label is inferred from the repository model-family name. Vision uses pipeline metadata or vision_config.
- license: Official source · verified 2026-10-22
- commercialUse: Official source · verified 2026-10-22. Official upstream organization metadata identifies a permissive standard license. Commercial use is permitted subject to the license terms, notices, and applicable restrictions; this is a metadata-based classification, not an individualized legal approval.
- license_url: Official source · verified 2026-10-22. No recognized standalone license file; linked pinned model card.
- architecture: Official source · verified 2026-10-22. Classification from model_type and expert/state-space configuration fields (granite); hybrid classification takes precedence over expert count.
- parametersB: Official source · verified 2026-10-22. Exact safetensors parameter total reported by the Hugging Face API, divided by 1e9.
- activeParametersB: Official source · verified 2026-10-22. Dense model: all parameters are included. This is not a runtime FLOP measurement.
- contextLength: Official source · verified 2026-10-22. Configured maximum token positions; not a guarantee of runtime support or quality at that length. No speculative RoPE scaling applied.
- kvCache: Official source · verified 2026-10-22. Standard attention KV shape from num_hidden_layers, num_key_value_heads, and head_dim or hidden_size / num_attention_heads; FP16 cache bytes are a separate estimator choice.
- licenseStandard: Official source · verified 2026-10-22. Official standard license text supporting the permissive-license classification; model applicability is identified separately by upstream metadata.