selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

Alternatives to RWKV7-G1k-13.3B-20260930

120 open-weight models that do the same job as RWKV7-G1k-13.3B-20260930 (13.3B chat), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.

the original
RWKV logo≈16 GB

This is an official BlinkDL release of RWKV-7 Goose in Hugging Face Transformers format. RWKV-7 is an attention-free recurrent architecture with a constant-size recurrent state and constant inference work per generated token. Training remains parallelizable.

13.3BApache 2.0
438Sept 2026
--vram
any
--gpus
--quant
  1. inclusionAI logo

    LLaDA2.0-mini

    inclusionAI·Chat

    ≈10 GB

    LLaDA2.0-mini is a diffusion language model featuring a 16BA1B Mixture-of-Experts (MoE) architecture. As an enhanced, instruction-tuned iteration of the LLaDA series, it is optimized for practical applications.

    16.3BA1.4B32K ctx Apache 2.0
    13.9KNov 2025
    • Same category
    • Similar size (16.3B vs 13.3B)
    • Commercial use OK
    vs
  2. allenai logo

    Emo_1b14b_1T

    allenai·Chat

    ≈9.0 GB

    The main release of EMO from EMO: Pretraining Mixture of Experts for Emergent Modularity — referred to as EMO (1T tokens, midtrained) in the paper.

    13.6BA1.5B4K ctx Apache 2.0
    736Apr 2026
    • Same category
    • Similar size (13.6B vs 13.3B)
    • Commercial use OK
    vs
  3. microsoft logo

    FrogMini-14B-2510

    microsoft·Chat

    ≈10 GB

    FrogMini is built on the Qwen3-14B transformer architecture with a maximum context length of 64k tokens. The model uses multi-turn debugging workflows and complex code reasoning. Unlike general-purpose LLMs, FrogMini is specialized for software engineering tasks.

    14B40K ctx MIT
    2.2KJan 2026gguf
    • Same category
    • Similar size (14B vs 13.3B)
    • Commercial use OK
    vs
  4. NousResearch logo

    Hermes-4-14B

    NousResearch·Chat

    ≈10 GB

    Hermes 4 14B is a frontier, hybrid-mode reasoning model based on Qwen 3 14B by Nous Research that is aligned to you.

    14B40K ctx Apache 2.0
    6.2KAug 2025gguf
    • Same category
    • Similar size (14B vs 13.3B)
    • Commercial use OK
    vs
  5. LiquidAI logo

    LFM2.5-24B-A2B

    LiquidAI·Chat

    ≈15 GB

    LFM2.5-24B-A2B is a 24B-parameter reasoning model with only 2B active parameters, built for fast, private agents on-device. It is the largest and final model in the LFM2.5 family, which now spans two orders of magnitude: from LFM2.5-230M to LFM2.5-24B-A2B.

    24BA2B128K ctx lfm1.0
    31.6KOct 2026gguf
    • Same category
    • Similar size (24B vs 13.3B)
    • Newer release
    vs
  6. ibm-granite logo

    granite-switch-4.1-8b

    ibm-granite·Chat

    ≈7.0 GB

    📝 Read the launch blog: Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

    8.8B128K ctx Apache 2.0
    6.2KOct 2026gguf
    • Same category
    • Similar size (8.8B vs 13.3B)
    • Commercial use OK
    • Newer release
    vs
  7. nvidia logo≈7.5 GB

    NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response.

    8.9B128K ctx NVIDIA Open Model License
    130.2KAug 2025gguf
    • Same category
    • Similar size (8.9B vs 13.3B)
    • Commercial use OK
    vs
  8. Qwen logo≈5.5 GB

    Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:

    7.6B32K ctx Apache 2.0
    6.5MSept 2024ollama
    • Same category
    • Similar size (7.6B vs 13.3B)
    • Commercial use OK
    • Widely used
    vs
  9. Qwen logo

    Qwen3-8B-Base

    Qwen·Chat

    ≈6.5 GB

    Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

    8.2B32K ctx Apache 2.0
    597.9KApr 2025gguf
    • Same category
    • Similar size (8.2B vs 13.3B)
    • Commercial use OK
    vs
  10. Qwen logo

    Qwen2.5-14B

    Qwen·Chat

    ≈11 GB

    Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:

    14.8B128K ctx Apache 2.0
    99.2KSept 2024gguf
    • Same category
    • Similar size (14.8B vs 13.3B)
    • Commercial use OK
    vs
  11. LiquidAI logo

    LFM2.5-8B-A1B

    LiquidAI·Chat

    ≈5.9 GB

    LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning.

    8.5BA1B125K ctx lfm1.0
    17KMay 2026ollama
    • Same category
    • Similar size (8.5B vs 13.3B)
    vs
  12. ornith-ai logo

    Ornith-1.0-35B

    ornith-ai·Chat

    ≈22 GB

    Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.

    35BA3.8B256K ctx MIT
    1.5MJun 2026gguf
    • Same category
    • Larger (35B vs 13.3B)
    • Commercial use OK
    • Widely used
    vs
  13. Vikhrmodels logo≈9.0 GB

    Vikhr-Nemo-12B-Instruct-R-21-09-24 is an open-weight chat model with 12.2B parameters from Vikhrmodels, published on Hugging Face.

    12.2B1000K ctx Apache 2.0
    59.8KSept 2024gguf
    • Same category
    • Similar size (12.2B vs 13.3B)
    • Commercial use OK
    vs
  14. meta-llama logo

    Llama-3.1-8B-Instruct

    meta-llama·Chat

    ≈6.3 GB

    The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out).

    8B128K ctx Llama 3.1 Community License
    4.2MJul 2024ollama
    • Same category
    • Similar size (8B vs 13.3B)
    • Widely used
    vs
  15. NVIDIA-Nemotron-Nano-12B-v2-Base is a large language model (LLM) developed by NVIDIA that is designed as a completion model for a given piece of text. It uses a hybrid model architecture that consists primarily of Mamba-2 and MLP layers with just six Attention layers.

    12.3BNVIDIA Open Model License
    4.1KAug 2025gguf
    • Same category
    • Similar size (12.3B vs 13.3B)
    • Commercial use OK
    vs
  16. swiss-ai logo≈6.3 GB

    Apertus is a 70B and 8B parameter language model designed to push the boundaries of fully-open multilingual and transparent models.

    8.1B64K ctx Apache 2.0
    204.9KAug 2025gguf
    • Same category
    • Similar size (8.1B vs 13.3B)
    • Commercial use OK
    vs
  17. speakleash logo

    Bielik-11B-v3.1-Instruct

    speakleash·Chat

    ≈8.3 GB

    Bielik-11B-v3.1-Instruct is a generative text model featuring 11 billion parameters. It is an update to the Bielik 11B v3 line that expands multilingual continual pre-training and strengthens post-training for instruction following, tool use, and reasoning.

    11.2B32K ctx Apache 2.0
    377Oct 2026gguf
    • Same category
    • Similar size (11.2B vs 13.3B)
    • Commercial use OK
    • Newer release
    vs
  18. inclusionAI logo

    Ling-3.0-tiny

    inclusionAI·Chat

    ≈6.7 GB

    We are introducing Ling-3.0-tiny, a lightweight hybrid reasoning MoE model with 7.9B total parameters and only 1.3B activated parameters per token.

    7.9BA1.4B128K ctx MIT
    13.6KAug 2026gguf
    • Same category
    • Similar size (7.9B vs 13.3B)
    • Commercial use OK
    vs
  19. nvidia logo≈9.1 GB

    NVIDIA-Nemotron-Nano-12B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response.

    12.3BNVIDIA Open Model License
    3.6KAug 2025gguf
    • Same category
    • Similar size (12.3B vs 13.3B)
    • Commercial use OK
    vs
  20. Qwen logo≈6.5 GB

    Qwen3Guard is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.

    8.2B32K ctx Apache 2.0
    109.4KSept 2025gguf
    • Same category
    • Similar size (8.2B vs 13.3B)
    • Commercial use OK
    vs
  21. allenai logo≈12 GB

    Upon the initial release of OLMo-2 models, we realized the post-trained models did not share the pre-tokenization logic that the base models use. As a result, we have trained new post-trained models.

    13.7B4K ctx Apache 2.0
    14.3KDec 2024ollama
    • Same category
    • Similar size (13.7B vs 13.3B)
    • Commercial use OK
    vs
  22. empero-ai logo≈7.1 GB

    Qwythos-9B is a full-parameter reasoning model built on top of a deeply uncensored Qwen3.5-9B base and post-trained on over 500 million tokens of high-quality Claude Mythos and Claude Fable traces, with chain-of-thought generated in-house by Empero AI's internal tool rethink.

    9.4B1M ctx Apache 2.0
    3.6KJun 2026gguf
    • Same category
    • Similar size (9.4B vs 13.3B)
    • Commercial use OK
    vs
  23. moonshotai logo≈12 GB

    Recently, the Muon optimizer has demonstrated strong results in training small-scale language models, but the scalability to larger models has not been proven. We identify two crucial techniques for scaling up Muon:

    16BA3B8K ctx MIT
    34.9KFeb 2025gguf
    • Same category
    • Similar size (16B vs 13.3B)
    • Commercial use OK
    vs
  24. zai-org logo

    GLM-4.7-Flash

    zai-org·Chat

    ≈26 GB

    👋 Join our Discord community. 📖 Check out the GLM-4.7 technical blog , technical report(GLM-4.5) . 📍 Use GLM-4.7-Flash API services on Z.ai API Platform. 👉 One click to GLM-4.7 .

    31.2BA4.6B198K ctx MIT
    1.1MJan 2026ollama
    • Same category
    • Larger (31.2B vs 13.3B)
    • Commercial use OK
    vs