ODS V3 Pre-Release: Public testing and refinement ahead of the official V3 launch. Turn your PC, Mac, or Linux box into a private AI server.
Model terms source inventory
Reviewed 2026-09-24 UTC against the 57 curated model records in ODS source
1bc5e1cbb24864f1c9efd551e0613202af324e3f. The qwen3.6-27b-ud-q4-k-xl
candidate row was added on 2026-09-25 with its catalog entry, for 58 records.
The gemma4-26b-a4b-q4 and gemma4-31b-q4 rows were re-pointed on 2026-09-25
to pinned unsloth revisions after ggml-org deleted their Q4_K_M files upstream.
The same day, every other direct catalog URL that requested main was pinned
to the revision listed here after its LFS sha256 at that revision matched the
catalog gguf_sha256; the terms metadata in those rows is unchanged.
The qwen3.8-flash-next-nvfp4-vllm and glm-5.3-flash-nvfp4-vllm rows were
added on 2026-10-27 with their catalog entries, for 60 records. They are vLLM
snapshots, not GGUFs: the catalog pins the whole repository revision and every
file's digest. The Qwen model card names the NVIDIA Open Model License for the
checkpoint and the Qwen Community License 1.0 for the base model.
This is a documentation inventory of upstream repository metadata, not a license grant, legal-compatibility decision, or completed per-artifact clearance. A quantizer's model-card tag can be incomplete or wrong and does not override the original model license. Each source link identifies the revision requested by the catalog; mutable refs need a separate immutable artifact receipt.
Four catalog entries have a top-level license (one of them also a
license_name), none has a top-level license_url, and one additional entry
has nested source_evidence license information. This page does not change that runtime schema or enforce acceptance.
The separate Hugging Face browser does display upstream license metadata; that
does not close the curated catalog's provenance and acceptance gaps.
Before redistribution or restricted use, validate the original author and model
revision, quantizer relationship, full license text, acceptable-use terms,
required notices, and any required acceptance against the actual artifact.
A missing or other tag requires review, not an assumption of Apache-2.0.
| Catalog model | Artifact-source metadata | Declared license tag | Original/base model metadata |
|---|---|---|---|
qwen3.5-2b-q4 |
unsloth/Qwen3.5-2B-GGUF @ f6d5376be1ed (requested f6d5376be1ed) |
apache-2.0 | ["Qwen/Qwen3.5-2B"] |
jamba-reasoning-3b-q4 |
ai21labs/AI21-Jamba-Reasoning-3B-GGUF @ 462e08a43c3c (requested 462e08a43c3c) |
apache-2.0 | Not declared |
phi4-mini-q4 |
unsloth/Phi-4-mini-instruct-GGUF @ 78eb92a46fc3 (requested 78eb92a46fc3) |
mit | microsoft/Phi-4-mini-instruct |
phi3.5-mini-q4 |
bartowski/Phi-3.5-mini-instruct-GGUF @ 6d70da17e749 (requested 6d70da17e749) |
mit | microsoft/Phi-3.5-mini-instruct |
phi4-mini-reasoning-q4 |
SandLogicTechnologies/Phi-4-mini-reasoning-GGUF @ 40e6071714bc (requested 40e6071714bc) |
mit | ["microsoft/Phi-4-mini-reasoning"] |
qwen2.5-1.5b-instruct-q4 |
Qwen/Qwen2.5-1.5B-Instruct-GGUF @ 91cad51170dc (requested 91cad51170dc) |
apache-2.0 | Qwen/Qwen2.5-1.5B-Instruct |
qwen2.5-0.5b-instruct-q4 |
Qwen/Qwen2.5-0.5B-Instruct-GGUF @ 9217f5db79a2 (requested 9217f5db79a2) |
apache-2.0 | Qwen/Qwen2.5-0.5B-Instruct |
granite3.3-2b-instruct-q4 |
unsloth/granite-3.3-2b-instruct-GGUF @ 8137ae9f7dd3 (requested 8137ae9f7dd3) |
Not declared | Not declared |
granite4.0-h-micro-q4 |
ibm-granite/granite-4.0-h-micro-GGUF @ dc1dd2585fac (requested dc1dd2585fac) |
apache-2.0 | ["ibm-granite/granite-4.0-h-micro"] |
granite4.0-h-tiny-q4 |
ibm-granite/granite-4.0-h-tiny-GGUF @ 08d5a8a9741d (requested 08d5a8a9741d) |
apache-2.0 | ["ibm-granite/granite-4.0-h-tiny"] |
smollm3-3b-q4 |
unsloth/SmolLM3-3B-GGUF @ a7bc17204c8a (requested a7bc17204c8a) |
apache-2.0 | ["HuggingFaceTB/SmolLM3-3B"] |
gemma3-4b-it-q4 |
ggml-org/gemma-3-4b-it-GGUF @ d09762237476 (requested d09762237476) |
gemma | ["google/gemma-3-4b-it"] |
granite4.0-h-1b-q4 |
ibm-granite/granite-4.0-h-1b-GGUF @ c2cb1972f511 (requested c2cb1972f511) |
apache-2.0 | ["ibm-granite/granite-4.0-h-1b"] |
falcon-h1-1.5b-instruct-q4 |
tiiuae/Falcon-H1-1.5B-Instruct-GGUF @ 0d3a6cfe25fb (requested 0d3a6cfe25fb) |
other | tiiuae/Falcon-H1-1.5B-Instruct |
falcon-h1-3b-instruct-q4 |
tiiuae/Falcon-H1-3B-Instruct-GGUF @ 18cc9812739f (requested 18cc9812739f) |
other | tiiuae/Falcon-H1-3B-Base |
nvidia-nemotron3-nano-4b-q4 |
nvidia/NVIDIA-Nemotron-3-Nano-4B-GGUF @ ba223d14e455 (requested ba223d14e455) |
other | nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8 |
granite4.1-3b-q4 |
ibm-granite/granite-4.1-3b-GGUF @ ab4701481089 (requested ab4701481089) |
apache-2.0 | ["ibm-granite/granite-4.1-3b"] |
granite4.0-1b-q4 |
ibm-granite/granite-4.0-1b-GGUF @ b27c2fe3f211 (requested b27c2fe3f211) |
apache-2.0 | ["ibm-granite/granite-4.0-1b"] |
granite4.0-h-350m-q4 |
ibm-granite/granite-4.0-h-350m-GGUF @ a864f823cce6 (requested a864f823cce6) |
apache-2.0 | ["ibm-granite/granite-4.0-h-350m"] |
granite3.2-2b-instruct-q4 |
ibm-research/granite-3.2-2b-instruct-GGUF @ 153d944aab8c (requested 153d944aab8c) |
apache-2.0 | ["ibm-granite/granite-3.2-2b-instruct"] |
granite3.1-2b-instruct-q4 |
bartowski/granite-3.1-2b-instruct-GGUF @ e47b8b46c04c (requested e47b8b46c04c) |
apache-2.0 | ibm-granite/granite-3.1-2b-instruct |
phi3-mini-128k-q4 |
QuantFactory/Phi-3-mini-128k-instruct-GGUF @ 0cc851bb3014 (requested 0cc851bb3014) |
mit | Not declared |
ministral3-8b-instruct-2512-q4 |
mistralai/Ministral-3-8B-Instruct-2512-GGUF @ 0102285ad796 (requested 0102285ad796) |
apache-2.0 | ["mistralai/Ministral-3-8B-Instruct-2512"] |
ministral-3b-instruct-q4 |
mradermacher/Ministral-3b-instruct-GGUF @ e31dc5d6827b (requested e31dc5d6827b) |
apache-2.0 | ministral/Ministral-3b-instruct |
llama3.2-1b-instruct-q4 |
unsloth/Llama-3.2-1B-Instruct-GGUF @ b69aef112e9f (requested b69aef112e9f) |
llama3.2 | meta-llama/Llama-3.2-1B-Instruct |
llama3.2-3b-instruct-q4 |
hugging-quants/Llama-3.2-3B-Instruct-Q4_K_M-GGUF @ eb72f2a08dd2 (requested eb72f2a08dd2) |
Not declared | meta-llama/Llama-3.2-3B-Instruct |
qwen2.5-3b-instruct-q4 |
Qwen/Qwen2.5-3B-Instruct-GGUF @ 7dabda4d13d5 (requested 7dabda4d13d5) |
other | Qwen/Qwen2.5-3B-Instruct |
qwen3-4b-q4 |
Qwen/Qwen3-4B-GGUF @ bc640142c66e (requested bc640142c66e) |
apache-2.0 | Qwen/Qwen3-4B |
qwen3-4b-instruct-2507-q4 |
unsloth/Qwen3-4B-Instruct-2507-GGUF @ a06e946bb6b6 (requested a06e946bb6b6) |
apache-2.0 | ["Qwen/Qwen3-4B-Instruct-2507"] |
qwen3-4b-128k-q4 |
unsloth/Qwen3-4B-128K-GGUF @ a21842b9582c (requested a21842b9582c) |
apache-2.0 | Qwen/Qwen3-4B |
qwen3-1.7b-q4 |
ggml-org/Qwen3-1.7B-GGUF @ daeb8e2d528a (requested daeb8e2d528a) |
apache-2.0 | Qwen/Qwen3-1.7B |
qwen2.5-coder-1.5b-128k-q4 |
unsloth/Qwen2.5-Coder-1.5B-Instruct-128K-GGUF @ 15e05dbeacf9 (requested 15e05dbeacf9) |
apache-2.0 | Qwen/Qwen2.5-Coder-1.5B-Instruct |
qwen2.5-coder-3b-128k-q4 |
unsloth/Qwen2.5-Coder-3B-Instruct-128K-GGUF @ 5326551926d0 (requested 5326551926d0) |
apache-2.0 | Qwen/Qwen2.5-Coder-3B-Instruct |
qwen2.5-7b-instruct-q4 |
No direct GGUF source | No direct Hugging Face GGUF source in this record | Not declared |
llama3.1-8b-instruct-q4 |
bartowski/Meta-Llama-3.1-8B-Instruct-GGUF @ bf5b95e96dac (requested bf5b95e96dac) |
llama3.1 | meta-llama/Meta-Llama-3.1-8B-Instruct |
granite3.3-8b-instruct-q4 |
ibm-granite/granite-3.3-8b-instruct-GGUF @ e40e9dd739c7 (requested e40e9dd739c7) |
apache-2.0 | ["ibm-granite/granite-3.3-8b-instruct"] |
mistral-nemo-12b-instruct-q4 |
QuantFactory/Mistral-Nemo-Instruct-2407-GGUF @ 6d01a4babf90 (requested 6d01a4babf90) |
apache-2.0 | Not declared |
qwen3.5-4b-q4 |
unsloth/Qwen3.5-4B-GGUF @ e87f176479d0 (requested e87f176479d0) |
apache-2.0 | ["Qwen/Qwen3.5-4B"] |
gemma4-e2b-q4 |
unsloth/gemma-4-E2B-it-GGUF @ 0314792d7f1f (requested 0314792d7f1f) |
apache-2.0 | google/gemma-4-E2B-it |
deepseek-r1-7b-q4 |
unsloth/DeepSeek-R1-Distill-Qwen-7B-GGUF @ 097680e4eed7 (requested 097680e4eed7) |
apache-2.0 | deepseek-ai/DeepSeek-R1-Distill-Qwen-7B |
gemma4-e4b-q4 |
unsloth/gemma-4-E4B-it-GGUF @ bfc15c382204 (requested bfc15c382204) |
apache-2.0 | google/gemma-4-E4B-it |
qwen3.5-9b-q4 |
unsloth/Qwen3.5-9B-GGUF @ 3885219b6810 (requested 3885219b6810) |
apache-2.0 | ["Qwen/Qwen3.5-9B"] |
phi4-q4 |
bartowski/phi-4-GGUF @ 19cd65f97c2f (requested 19cd65f97c2f) |
mit | microsoft/phi-4 |
deepseek-r1-14b-q4 |
unsloth/DeepSeek-R1-Distill-Qwen-14B-GGUF @ 7b05b58b41f6 (requested 7b05b58b41f6) |
apache-2.0 | deepseek-ai/DeepSeek-R1-Distill-Qwen-14B |
qwen3.8-27b-iq4-xs |
unsloth/Qwen3.8-27B-GGUF @ 4ca720788d1e (requested 4ca720788d1e) |
apache-2.0 | ["Qwen/Qwen3.8-27B"] |
qwen3.5-27b-q4 |
unsloth/Qwen3.5-27B-GGUF @ 3221f178a6b8 (requested 3221f178a6b8) |
apache-2.0 | ["Qwen/Qwen3.5-27B"] |
qwen3.6-27b-ud-q4-k-xl |
unsloth/Qwen3.6-27B-GGUF @ 82d411acf4a0 (requested 82d411acf4a0) |
apache-2.0 | ["Qwen/Qwen3.6-27B"] |
gemma4-26b-a4b-q4 |
unsloth/gemma-4-26B-A4B-it-GGUF @ c099eb48e663 (requested c099eb48e663) |
apache-2.0 | google/gemma-4-26B-A4B-it |
qwen3-30b-a3b-q4 |
unsloth/Qwen3-30B-A3B-GGUF @ d5b1d57bd0b5 (requested d5b1d57bd0b5) |
apache-2.0 | Qwen/Qwen3-30B-A3B |
gemma4-31b-q4 |
unsloth/gemma-4-31B-it-GGUF @ c1ac76e99d55 (requested c1ac76e99d55) |
apache-2.0 | google/gemma-4-31B-it |
deepseek-r1-32b-q4 |
unsloth/DeepSeek-R1-Distill-Qwen-32B-GGUF @ 1938d05cc893 (requested 1938d05cc893) |
apache-2.0 | deepseek-ai/DeepSeek-R1-Distill-Qwen-32B |
qwen3.5-35b-a3b-q4 |
unsloth/Qwen3.5-35B-A3B-GGUF @ bc014a17be43 (requested bc014a17be43) |
apache-2.0 | ["Qwen/Qwen3.5-35B-A3B"] |
qwen3.6-35b-a3b-ud-q4 |
unsloth/Qwen3.6-35B-A3B-GGUF @ a483e9e6cbd5 (requested a483e9e6cbd5) |
apache-2.0 | ["Qwen/Qwen3.6-35B-A3B"] |
kat-coder-v2.5-dev-apex-q4 |
mudler/KAT-Coder-V2.5-Dev-APEX-GGUF @ be23ff3a49ee (requested be23ff3a49ee) |
apache-2.0 | Kwaipilot/KAT-Coder-V2.5-Dev |
deepseek-r1-70b-q4 |
unsloth/DeepSeek-R1-Distill-Llama-70B-GGUF @ 732dd974083e (requested 732dd974083e) |
llama3.3 | deepseek-ai/DeepSeek-R1-Distill-Llama-70B |
qwen3-coder-next-q4 |
unsloth/Qwen3-Coder-Next-GGUF @ ce09c67b53bc (requested ce09c67b53bc) |
apache-2.0 | ["Qwen/Qwen3-Coder-Next"] |
llama4-scout-q4 |
No direct GGUF source | No direct Hugging Face GGUF source in this record | Not declared |
qwen3.5-122b-a10b-q4 |
No direct GGUF source | No direct Hugging Face GGUF source in this record | Not declared |
qwen3.8-flash-next-nvfp4-vllm |
nvidia/Qwen3.8-Flash-Next-NVFP4 @ fc694b54fb01 (requested fc694b54fb01; vLLM snapshot) |
other (nvidia-open-model-license) |
["Qwen/Qwen3.8-Flash-Next"] |
glm-5.3-flash-nvfp4-vllm |
nvidia/GLM-5.3-Flash-NVFP4 @ da920bb0b9f4 (requested da920bb0b9f4; vLLM snapshot) |
mit | ["zai-org/GLM-5.3-Flash"] |
Maintenance
Rebuild this evidence when an artifact source or revision changes. Keep this inventory distinct from the runtime catalog until the runtime terms schema and acceptance flow are separately reviewed and tested. See third-party licensing for remaining work.