MCP Agent Tools
Open-source MCP servers, agent frameworks, eval harnesses, and tool-use projects for agent builders.
Focused lists make the catalog useful immediately while mirror hosting remains selective and reviewed.
- Multi-Source AI Popular non-Hugging Face records from ModelScope, GitHub, arXiv, OpenML, Zenodo, Dataverse, DataCite, and other indexed open-source AI sources.
- New Open Source Recently updated public open AI artifacts across every indexed source.
How this ranks. The order uses the recorded signals described above. Runtime and verification details live on each model’s record. Self-reported evidence has no trust or ranking impact.
Mirror candidate means the files still live upstream, not on our servers yet. Hosted files do not establish a complete runnable model. Download counts, where recorded, are downloads on Hugging Face.
Showing stale catalog ·
| Type | Model | Recorded size | Downloads | Updated |
|---|---|---|---|---|
ToolACE 8BTeam-ACE/ToolACE-8BThe catalog does not give enough information to describe what this model is for. Catalog sources: license.Content controls: not assessed.DetailsThe catalog does not give enough information to describe what this model is for.See the publisher’s description →Source: Team-ACE on Hugging Face. File format: Safetensors.
| 16.1 GB | 77,864 downloads on Hugging Face | 2025-08-20 | |
Parable Granite 4.1 3B Claude Fable 5AnkitAI/Parable-Granite-4.1-3B-Claude-Fable-5-GGUFThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.Source: AnkitAI on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 2.1 GB | 380,056 downloads on Hugging Face | 2026-09-16 | |
Llama xLAM 2 8B fc rSalesforce/Llama-xLAM-2-8b-fc-rThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.Source: Salesforce on Hugging Face. File format: Safetensors.
| 16.1 GB | 32,714 downloads on Hugging Face | 2025-05-06 | |
NeoHorse 1 4BTokenRhythm/NeoHorse-1-4B-GGUFThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: TokenRhythm on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 2.71 GB | 24,753 downloads on Hugging Face | 2026-09-10 | |
NeoHorse 1 9BTokenRhythm/NeoHorse-1-9B-GGUFThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: TokenRhythm on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 5.63 GB | 23,672 downloads on Hugging Face | 2026-09-20 | |
phonellm alpha 1EryriLabs/phonellm-alpha-1-GGUFThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.Source: EryriLabs on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 24.5 GB | 22,417 downloads on Hugging Face | 2026-09-21 | |
Parable Qwen3 8B Claude Fable 5AnkitAI/Parable-Qwen3-8B-Claude-Fable-5-GGUFThe catalog lists this model for generating text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text.Source: AnkitAI on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 5.03 GB | 595,751 downloads on Hugging Face | 2026-09-16 | |
Parable Granite 4.1 8B Claude Fable 5AnkitAI/Parable-Granite-4.1-8B-Claude-Fable-5-GGUFThe catalog lists this model for generating text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text.Source: AnkitAI on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 5.12 GB | 375,773 downloads on Hugging Face | 2026-09-16 | |
Agents A1 4B Q4 K MInternScience/Agents-A1-4B-Q4_K_M-GGUFThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: InternScience on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 3.38 GB | 186,394 downloads on Hugging Face | 2026-07-14 | |
dolphin 2.9 llama3 8Bbartowski/dolphin-2.9-llama3-8b-GGUFThe catalog lists this model for having conversations. Catalog sources: tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 4.92 GB | 19,626 downloads on Hugging Face | 2024-10-31 | |
cognitivecomputations Dolphin3.0 R1 Mistral 24Bbartowski/cognitivecomputations_Dolphin3.0-R1-Mistral-24B-GGUFThe catalog lists this model for generating text. Catalog sources: task.Content controls: not assessed.DetailsThe catalog lists this model for generating text.Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 14.3 GB | 15,281 downloads on Hugging Face | 2026-09-18 | |
Agents A1 4BInternScience/Agents-A1-4BThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: InternScience on Hugging Face. File format: Safetensors.
| 9.1 GB | 343,675 downloads on Hugging Face | 2026-07-14 | |
multilingual e5 small aligned sentimentagentlans/multilingual-e5-small-aligned-sentimentThe catalog lists this model for sorting text into categories. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for sorting text into categories.Source: agentlans on Hugging Face. File format: Safetensors, BIN.
| 492.9 MB | 21,388 downloads on Hugging Face | 2024-12-15 | |
Parable Qwen3 4B Claude Fable 5AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-GGUFThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: AnkitAI on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 2.5 GB | 495,672 downloads on Hugging Face | 2026-09-16 | |
GLM 4.7 Flash Coderwhitecircle/GLM-4.7-Flash-CoderThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: whitecircle on Hugging Face. File format: Safetensors.
| 59.9 GB | 24,865 downloads on Hugging Face | 2026-09-12 | |
Dolphin3.0 Llama3.1 8Bdphn/Dolphin3.0-Llama3.1-8B-GGUFThe catalog does not give enough information to describe what this model is for. Catalog sources: license.Content controls: not assessed.DetailsThe catalog does not give enough information to describe what this model is for.See the publisher’s description →Source: dphn on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 4.92 GB | 43,013 downloads on Hugging Face | 2026-09-18 | |
Dolphin3.0 Llama3.1 8Bdphn/Dolphin3.0-Llama3.1-8BThe catalog does not give enough information to describe what this model is for. Catalog sources: license.Content controls: not assessed.DetailsThe catalog does not give enough information to describe what this model is for.See the publisher’s description →Source: dphn on Hugging Face. File format: Safetensors.
| 16.1 GB | 35,703 downloads on Hugging Face | 2026-09-20 | |
Qwen3.6 35B A3B DFlash GGUF llama.cppAlittlehammmer/Qwen3.6-35B-A3B-DFlash-GGUF-llama.cppThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: Alittlehammmer on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 2.05 GB | 75,433 downloads on Hugging Face | 2026-09-16 | |
Qwen3.6 27B DFlash GGUF llama.cppAlittlehammmer/Qwen3.6-27B-DFlash-GGUF-llama.cppThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: Alittlehammmer on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 1.03 GB | 73,764 downloads on Hugging Face | 2026-09-16 | |
NVIDIA Nemotron 3 Nano 30B A3B MLX 4bitlmstudio-community/NVIDIA-Nemotron-3-Nano-30B-A3B-MLX-4bitThe catalog lists this model for having conversations. Catalog sources: tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: lmstudio-community on Hugging Face. File format: Safetensors.
| 17.8 GB | 46,134 downloads on Hugging Face | 2026-09-18 | |
Nemotron 3 Nano 30B A3Bunsloth/Nemotron-3-Nano-30B-A3B-GGUFThe catalog lists this model for generating text. Catalog sources: task.Content controls: not assessed.DetailsThe catalog lists this model for generating text.Source: unsloth on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 24.6 GB | 14,911 downloads on Hugging Face | 2026-09-18 | |
Qwen3.8 14B Turbomradermacher/Qwen3.8-14B-Instruct-Turbo-GGUFThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.See the publisher’s description →Source: mradermacher on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 103.3 GB | 17,784 downloads on Hugging Face | 2026-08-29 | |
supergemma4 26B uncensored gguf v2Jiunsong/supergemma4-26b-uncensored-gguf-v2The catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: Jiunsong on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 16.8 GB | 335,245 downloads on Hugging Face | 2026-09-18 | |
SignOfFour i1mradermacher/SignOfFour-i1-GGUFThe catalog does not give enough information to describe what this model is for. Catalog sources: license.Content controls: not assessed.DetailsThe catalog does not give enough information to describe what this model is for.See the publisher’s description →Source: mradermacher on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 21.2 GB | 18,502 downloads on Hugging Face | 2026-09-07 | |
Qwen3.6 14B A3B FableVibestvall43/Qwen3.6-14B-A3B-FableVibes-GGUFThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: tvall43 on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 75 GB | 34,342 downloads on Hugging Face | 2026-06-15 | |
granite 4.2 30Bbartowski/granite-4.2-30b-GGUFThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 18 GB | 14,670 downloads on Hugging Face | 2026-08-25 | |
Qwen3.8 27B EfficientThink Uncensored K3 Opus5 Grok4.6 GPT5.6Sol SFT SimPO DFlash2nerkyor/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2The catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: nerkyor on Hugging Face. File format: Safetensors.
| 433.6 GB | 36,442 downloads on Hugging Face | 2026-09-12 | |
Qwen3.6 27B DSV4Pro Thinking Distillnerkyor/Qwen3.6-27B-DSV4Pro-Thinking-DistillThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: nerkyor on Hugging Face. File format: Safetensors, GGUF. GGUF files use llama.cpp-compatible runners.
| 292 GB | 16,730 downloads on Hugging Face | 2026-06-28 | |
DeepSeek V4 Pro Qwen3.5 9B MTPJackrong/DeepSeek-V4-Pro-Qwen3.5-9B-MTP-GGUFThe catalog lists this model for asking software tools to take actions. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for asking software tools to take actions.Source: Jackrong on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 85.4 GB | 36,137 downloads on Hugging Face | 2026-08-13 | |
JiRackUltra 1BCMSManhattan/JiRackUltra_1bThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: CMSManhattan on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 16.1 GB | 1,883,610 downloads on Hugging Face | 2026-10-04 | |
Qwopus3.6 35B A3B Coder MTPJackrong/Qwopus3.6-35B-A3B-Coder-MTP-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: Jackrong on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 133 GB | 469,734 downloads on Hugging Face | 2026-07-09 | |
Qwen3.8 27B OrcaRouter GSQ RCO IQ3 XXS UncensoredRentedNoodle/Qwen3.8-27B-OrcaRouter-GSQ-RCO-IQ3_XXS-UncensoredThe catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: RentedNoodle on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 32.9 GB | 45,917 downloads on Hugging Face | 2026-10-03 | |
mmBERT basejhu-clsp/mmBERT-baseThe catalog lists this model for filling in missing words in text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for filling in missing words in text.Source: jhu-clsp on Hugging Face. File format: BIN.
| 1.25 GB | 358,921 downloads on Hugging Face | 2025-10-07 | |
stanford deidentifier baseStanfordAIMI/stanford-deidentifier-baseThe catalog lists this model for labeling parts of text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for labeling parts of text.Source: StanfordAIMI on Hugging Face. File format: BIN.
| 438.3 MB | 1,229,979 downloads on Hugging Face | 2024-10-09 | |
BAAI AREX Turbobartowski/BAAI_AREX-Turbo-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 73.9 GB | 227,939 downloads on Hugging Face | 2026-07-25 | |
mmBERT smalljhu-clsp/mmBERT-smallThe catalog lists this model for filling in missing words in text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for filling in missing words in text.Source: jhu-clsp on Hugging Face. File format: BIN.
| 581.2 MB | 21,679 downloads on Hugging Face | 2025-10-17 | |
Qwable 9B Claude Fable 5empero-ai/Qwable-9B-Claude-Fable-5The catalog lists this model for having conversations. Catalog sources: license, tags, task.Content controls: not assessed.DetailsThe catalog lists this model for having conversations.Source: empero-ai on Hugging Face. File format: Safetensors.
| 18.8 GB | 26,403 downloads on Hugging Face | 2026-06-19 | |
Tiel Coder 35B A3Bpeculiar-ragdoll/Tiel-Coder-35B-A3B-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: peculiar-ragdoll on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 202.8 GB | 16,011 downloads on Hugging Face | 2026-09-10 | |
Tiel Coder 35B A3B GGUF MTPpeculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTPThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: peculiar-ragdoll on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 204.6 GB | 13,959 downloads on Hugging Face | 2026-09-10 | |
Qwythos 9B Claude Mythos 5 1Mempero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: empero-ai on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 96.2 GB | 907,682 downloads on Hugging Face | 2026-07-14 | |
MiniMax M3MiniMaxAI/MiniMax-M3The catalog lists this model for generating text from images and text. Catalog sources: task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.Source: MiniMaxAI on Hugging Face.
| 854.2 GB | 171,422 downloads on Hugging Face | 2026-07-23 | |
XYZAILab XYZ Aquila Minibartowski/XYZAILab_XYZ-Aquila-mini-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 570.8 GB | 883,330 downloads on Hugging Face | 2026-07-28 | |
Gemma4 12B QAT Uncensored HauhauCS BalancedHauhauCS/Gemma4-12B-QAT-Uncensored-HauhauCS-BalancedThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: HauhauCS on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 7.38 GB | 626,923 downloads on Hugging Face | 2026-06-25 | |
Fara1.5 4Bbartowski/Fara1.5-4B-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 73.9 GB | 239,821 downloads on Hugging Face | 2026-07-22 | |
tencent UI Mate 9Bbartowski/tencent_UI-Mate-9B-GGUFThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: bartowski on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 153.5 GB | 225,858 downloads on Hugging Face | 2026-08-18 | |
Qwen3.6 27B Uncensored HauhauCS BalancedHauhauCS/Qwen3.6-27B-Uncensored-HauhauCS-BalancedThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: HauhauCS on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 169.8 GB | 92,038 downloads on Hugging Face | 2026-04-23 | |
Qwen3.6 35B A3B NVFP4sakamakismile/Qwen3.6-35B-A3B-NVFP4The catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: sakamakismile on Hugging Face. File format: Safetensors.
| 21.9 GB | 59,376 downloads on Hugging Face | 2026-04-29 | |
Smaug Miniabacusai/Smaug-MiniThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: abacusai on Hugging Face. File format: Safetensors.
| 55.6 GB | 18,308 downloads on Hugging Face | 2026-09-09 | |
Huihui Qwen3.6 35B A3B abliterated NVFP4sakamakismile/Huihui-Qwen3.6-35B-A3B-abliterated-NVFP4The catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: uploader describes removal; not independently verified.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: sakamakismile on Hugging Face. File format: Safetensors.
| 21.9 GB | 14,466 downloads on Hugging Face | 2026-04-29 | |
Qwen3.8 27B GSQ RCO IQ3 XXS UncensoredRentedNoodle/Qwen3.8-27B-GSQ-RCO-IQ3_XXS-UncensoredThe catalog lists this model for generating text from images and text. Catalog sources: license, task.Content controls: not assessed.DetailsThe catalog lists this model for generating text from images and text.See the publisher’s description →Source: RentedNoodle on Hugging Face. File format: GGUF. GGUF files use llama.cpp-compatible runners.
| 22.5 GB | 32,805 downloads on Hugging Face | 2026-09-07 |