# Text Generation Models, Page 21: Open-Weight AI Source: https://savrn.com/models/tasks/text-generation/page/21 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- SAVRN Model Hub · Models by Task # Text Generation Models 1,548 open-weight text generation models in the SAVRN Model Hub, with Convergent Intelligence, Qwen and Francesca Padovani publishing the most. 1,548Models 721Publishers 107K to 1.7TParameter range 8Licenses ## SAVRN's Take Text generation is where most of the attention goes, and where most of the money is spent. The models on this page run from tiny research checkpoints that fit on a laptop to frontier-class systems that need a full server of accelerators. The first question we ask about any of them is not how smart it is, but what it costs to keep it answering around the clock. That cost comes down to three things: the size of the weights, the precision you run them at, and how much context each request carries. Moving from 16-bit to 4-bit cuts the memory a model needs by roughly four times, which is often the difference between one server and several. Mixture-of-experts models, such as the newer DeepSeek, Qwen and Nemotron releases, hold a large total parameter count but activate only a small share per token, so they answer faster than their size suggests while still needing the memory to hold every expert. Licensing decides the rest. Apache 2.0 and MIT models can go straight into a commercial product. Llama, Gemma's earlier generations and models marked other each carry terms you need to read first, and a model with no stated license is research material until the publisher says otherwise. Our advice is to shortlist by license, size to the hardware you can actually run, then test the survivors on your own prompts. The table below is sorted by what people download, not by what we recommend. SAVRN Research, 2026-09-18 ## Most Downloaded | Model | Publisher | Parameters | License | Monthly downloads | Cheapest GPUs at 16-bit | | --- | --- | --- | --- | --- | --- | | [Qwen3-0.6B](https://savrn.com/models/qwen3-0-6b) | Qwen | 752M | apache-2.0 | 26.7M | 1x MI300X, $1.85/hr | | [gpt2](https://savrn.com/models/gpt2) | OpenAI community | 137M | mit | 18.6M | 1x MI300X, $1.85/hr | | [Qwen3-4B](https://savrn.com/models/qwen3-4b) | Qwen | 4B | apache-2.0 | 11.6M | 1x MI300X, $1.85/hr | | [Qwen3-14B](https://savrn.com/models/qwen3-14b) | Qwen | 14.8B | apache-2.0 | 9.2M | 1x MI300X, $1.85/hr | | [Qwen3-8B](https://savrn.com/models/qwen3-8b) | Qwen | 8.2B | apache-2.0 | 8.3M | 1x MI300X, $1.85/hr | | [Qwen2.5-7B-Instruct](https://savrn.com/models/qwen2-5-7b-instruct) | Qwen | 7.6B | apache-2.0 | 6.6M | 1x MI300X, $1.85/hr | | [gpt-oss-20b](https://savrn.com/models/gpt-oss-20b) | OpenAI | 20.9B | apache-2.0 | 5.8M | 1x MI300X, $1.85/hr | | [opt-125m](https://savrn.com/models/opt-125m) | AI at Meta | — | other | 5.8M | — | | [Qwen2.5-0.5B-Instruct](https://savrn.com/models/qwen2-5-0-5b-instruct) | Qwen | 494M | apache-2.0 | 5.5M | 1x MI300X, $1.85/hr | | [DeepSeek-V3.2](https://savrn.com/models/deepseek-v3-2) | DeepSeek | 685.4B | mit | 5.5M | 7x MI325X, $14.00/hr | ## Licenses | License | Models | Commercial use | | --- | --- | --- | | [apache-2.0](https://savrn.com/models/licenses/apache-2-0) | 714 | Yes | | not stated | 341 | Not stated | | other | 168 | Read the license | | [mit](https://savrn.com/models/licenses/mit) | 151 | Yes | | [cc-by-nc-4.0](https://savrn.com/models/licenses/cc-by-nc-4-0) | 41 | Not without separate permission | | [llama3.2](https://savrn.com/models/licenses/llama3-2) | 33 | Read the license | | [gemma](https://savrn.com/models/licenses/gemma) | 22 | Yes, with conditions | | [llama3.1](https://savrn.com/models/licenses/llama3-1) | 21 | Yes, with conditions | ## Who Publishes Them | Publisher | Models | | --- | --- | | [Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow) | 54 | | [Qwen](https://savrn.com/model-publishers/qwen) | 42 | | [Francesca Padovani](https://savrn.com/model-publishers/francesca9805) | 37 | | [Yanggangu](https://savrn.com/model-publishers/yanggangu) | 28 | | [Shreyansh singh](https://savrn.com/model-publishers/shreyansh12183) | 25 | | [Kumar](https://savrn.com/model-publishers/sandeep123) | 23 | ## All 1,548 Models, Page 21 of 26 Model · Text generation ### [Qwen3-32B-target-only-no-hallucination-first-third-sft-bf16](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-first-third-sft-bf16) [Localized Ft](https://savrn.com/model-publishers/localized-ft) This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library. Open weights apache-2.0 40,960 tokens transformers [View model](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-first-third-sft-bf16) Model · Text generation ### [Qwen3-32B-target-only-no-hallucination-last-third-sft-bf16](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-last-third-sft-bf16) [Localized Ft](https://savrn.com/model-publishers/localized-ft) This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library. Open weights apache-2.0 32.8B parameters 40,960 tokens transformers [View model](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-last-third-sft-bf16) Model · Text generation ### [Qwen3-32B-target-only-no-hallucination-second-third-sft-bf16](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-second-third-sft-bf16) [Localized Ft](https://savrn.com/model-publishers/localized-ft) This qwen3 model was trained 2x faster with Unsloth and Huggingface's TRL library. Open weights apache-2.0 32.8B parameters 40,960 tokens transformers [View model](https://savrn.com/models/qwen3-32b-target-only-no-hallucination-second-third-sft-bf16) Model · Text generation ### [qwen3_4b_8b_abliterated_fp16_convrot_int8](https://savrn.com/models/qwen3-4b-8b-abliterated-fp16-convrot-int8) [GJL](https://savrn.com/model-publishers/ussoewwin) Open weights apache-2.0 transformers [View model](https://savrn.com/models/qwen3-4b-8b-abliterated-fp16-convrot-int8) Model · Text generation ### [qwen3-4b-ablation-poe-no-base](https://savrn.com/models/qwen3-4b-ablation-poe-no-base) [Op](https://savrn.com/model-publishers/op12138) The full ImmuneCoT method fuses the two safety branches with a Base-adjusted product-of-experts Qimm(v) ∝ q̃rec(v)·q̃res(v)/qB(v). This checkpoint uses the naive fusion Qno-base(v) ∝ q̃rec(v)·q̃res(v) — the same branch weights (λrec=0.5, λres=0.7) but no division by the base distribution — isolating whether the gains come from combining Recognition+Response at all, or specifically from the Base-adjusted PoE term. Intended use: research reproducibility for the ImmuneCoT paper's RQ3 ablation. Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-ablation-poe-no-base) Model · Text generation ### [Qwen3-4B-expert-Code](https://savrn.com/models/qwen3-4b-expert-code) [David S](https://savrn.com/model-publishers/david260928) Code generation expert. Full BF16 weights with configuration and tokenizer. Supervised fine-tuned Qwen3 weights. Apache 2.0; see LICENSE and NOTICE. Open weights apache-2.0 4B parameters 65,536 tokens transformers [View model](https://savrn.com/models/qwen3-4b-expert-code) Model · Text generation ### [Qwen3-4B-expert-Math](https://savrn.com/models/qwen3-4b-expert-math) [David S](https://savrn.com/model-publishers/david260928) Mathematical reasoning expert. Full BF16 weights with configuration and tokenizer. Supervised fine-tuned Qwen3 weights. Apache 2.0; see LICENSE and NOTICE. Open weights apache-2.0 4B parameters 65,536 tokens transformers [View model](https://savrn.com/models/qwen3-4b-expert-math) Model · Text generation ### [Qwen3-4B-Instruct-2507-abliterated](https://savrn.com/models/qwen3-4b-instruct-2507-abliterated) [Seva](https://savrn.com/model-publishers/glivorem) An abliterated version of Qwen/Qwen3-4B-Instruct-2507, made with Heretic. Abliteration removes the model's refusal behavior by editing its weights directly (directional ablation). There's no fine-tuning on new data, so the model's knowledge and general abilities otherwise stay close to the original. Works anywhere the base model does. With vLLM: Recommended sampling (from the base model): temperature=0.7, topp=0.8, topk=20. This model will comply with requests the original model would refuse. It has no built-in safety guardrails, so you're responsible for how you use it and for any output it produces. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-instruct-2507-abliterated) Model · Text generation ### [Qwen3-4B-JNS](https://savrn.com/models/qwen3-4b-jns) [Prabanta Dev](https://savrn.com/model-publishers/prabanta-dev) On a laptop running Debian 13 Linux, with an Intel Core Ultra 9 185H, its integrated Arc GPU and 32 GB of memory, Janas-LLM reads a prompt at 193 tok/s and writes at 27 tok/s: llama.cpp, on the same laptop, 164 and 18. The other files of this repository, measured the same way: Measured on 9-11 October 2026: llama-bench -p 1024 -n 16 and janas-bench --prompt 1024 --gen 16, both at their best configuration; the draft pair in a chat reply at the chat's own temperature, as the README's section on --draft tells (a draft guesses the next tokens and the model checks them: the same text, faster). On other machines the figures will differ. Details: Janas README. - Debian 13 (trixie) Linux, x86-64… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-4b-jns) Model · Text generation ### [qwen3-4b-medicine-online-pilot-261017-seed42-step-045](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-045) [[HYU_NLP] EVA TEAM2](https://savrn.com/model-publishers/hyu-nlp-eva2) Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-045) Model · Text generation ### [qwen3-4b-medicine-online-pilot-261017-seed42-step-050](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-050) [[HYU_NLP] EVA TEAM2](https://savrn.com/model-publishers/hyu-nlp-eva2) Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-050) Model · Text generation ### [qwen3-4b-medicine-online-pilot-261017-seed42-step-055](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-055) [[HYU_NLP] EVA TEAM2](https://savrn.com/model-publishers/hyu-nlp-eva2) Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-055) Model · Text generation ### [qwen3-4b-medicine-online-pilot-261017-seed42-step-060](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-060) [[HYU_NLP] EVA TEAM2](https://savrn.com/model-publishers/hyu-nlp-eva2) Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-060) Model · Text generation ### [qwen3-4b-medicine-online-pilot-261017-seed42-step-065](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-065) [[HYU_NLP] EVA TEAM2](https://savrn.com/model-publishers/hyu-nlp-eva2) Open weights apache-2.0 4B parameters transformers [View model](https://savrn.com/models/qwen3-4b-medicine-online-pilot-261017-seed42-step-065) Model · Text generation ### [qwen3-4b-medicine-static-pilot-261016-step-041](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-041) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) The root contains a BF16 Transformers export. originalcheckpoint/ contains the exact native veRL/FSDP checkpoint files and metadata. The local checkpoint is removed only after all remote file sizes and SHA256 hashes are verified. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-041) Model · Text generation ### [qwen3-4b-medicine-static-pilot-261016-step-042](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-042) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) The root contains a BF16 Transformers export. originalcheckpoint/ contains the exact native veRL/FSDP checkpoint files and metadata. The local checkpoint is removed only after all remote file sizes and SHA256 hashes are verified. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-042) Model · Text generation ### [qwen3-4b-medicine-static-pilot-261016-step-043](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-043) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) The root contains a BF16 Transformers export. originalcheckpoint/ contains the exact native veRL/FSDP checkpoint files and metadata. The local checkpoint is removed only after all remote file sizes and SHA256 hashes are verified. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-043) Model · Text generation ### [qwen3-4b-medicine-static-pilot-261016-step-044](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-044) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) The root contains a BF16 Transformers export. originalcheckpoint/ contains the exact native veRL/FSDP checkpoint files and metadata. The local checkpoint is removed only after all remote file sizes and SHA256 hashes are verified. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-medicine-static-pilot-261016-step-044) Model · Text generation ### [qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-003](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-003) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-online-rubrics-medicine-full-dense-20260919-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-003) Model · Text generation ### [qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-004](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-004) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-online-rubrics-medicine-full-dense-20260919-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-004) Model · Text generation ### [qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-005](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-005) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-online-rubrics-medicine-full-dense-20260919-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-005) Model · Text generation ### [qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-006](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-006) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-online-rubrics-medicine-full-dense-20260919-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-onlinerubrics-dense-seed11-step-006) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-022](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-022) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-static-r0-medicine-qwen3-4b-matched-dense-20260928-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-022) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-023](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-023) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-static-r0-medicine-qwen3-4b-matched-dense-20260928-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-023) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-024](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-024) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-static-r0-medicine-qwen3-4b-matched-dense-20260928-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-024) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-025](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-025) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-static-r0-medicine-qwen3-4b-matched-dense-20260928-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-025) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-026](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-026) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) Run: phase1-static-r0-medicine-qwen3-4b-matched-dense-20260928-seed11. The root contains a BF16 model for inference. originalcheckpoint/ contains the original veRL checkpoint files (model parameters only). The local checkpoint remains on disk. Research use only. Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-dense-seed11-step-026) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-seed11-step-000](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-000) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) This is the policy after 0 global optimizer updates of the matched separate from the OnlineRubrics/dynamic-rubric checkpoints. The root files are a BF16 Transformers export for inference. The originalcheckpoint/ directory contains the exact original veRL/FSDP policy parameter checkpoint and its tokenizer/configuration files. Optimizer, trainer, and data-loader state are intentionally not published; the complete resume checkpoint remains on Daisy. This is an intermediate research checkpoint, not a clinical model. No medical capability or safety claim is made. Original actor parameter SHA256: f81409edc253a52ee9b3e6807bf280cf1ff242c77c2f6645b087c74e03a4e3d4 Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-000) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-seed11-step-003](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-003) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) This is the policy after 3 global optimizer updates of the matched separate from the OnlineRubrics/dynamic-rubric checkpoints. The root files are a BF16 Transformers export for inference. The originalcheckpoint/ directory contains the exact original veRL/FSDP policy parameter checkpoint and its tokenizer/configuration files. Optimizer, trainer, and data-loader state are intentionally not published; the complete resume checkpoint remains on Daisy. This is an intermediate research checkpoint, not a clinical model. No medical capability or safety claim is made. Original actor parameter SHA256: 3339fe6d830916167c26f9e818cef25d0cecdbf3ab26e8545fcbcbf6465bfdc8 Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-003) Model · Text generation ### [qwen3-4b-rar-medicine-static-r0-matched-seed11-step-042](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-042) [[HYU_NLP] EVA Team](https://savrn.com/model-publishers/hyu-nlp-eval) This is the policy after 42 global optimizer updates of the matched separate from the OnlineRubrics/dynamic-rubric checkpoints. The root files are a BF16 Transformers export for inference. The originalcheckpoint/ directory contains the exact original veRL/FSDP policy parameter checkpoint and its tokenizer/configuration files. Optimizer, trainer, and data-loader state are intentionally not published; the complete resume checkpoint remains on Daisy. This is an intermediate research checkpoint, not a clinical model. No medical capability or safety claim is made. Original actor parameter SHA256: 3c024ab64a34d4140f446262e7fcee7fbd64da5b9de22cac486a986288e28063 Open weights apache-2.0 4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-4b-rar-medicine-static-r0-matched-seed11-step-042) Model · Text generation ### [qwen3-4b-tools-v1](https://savrn.com/models/qwen3-4b-tools-v1) [Abhijit Kapure](https://savrn.com/model-publishers/abhijitk20) A LoRA fine-tune of unsloth/Qwen3-4B that teaches the model when not to call a tool. On BFCL v4 Irrelevance Detection, under identical conditions and a full 240-entry category: The baselines are statistically indistinguishable from this adapter. v1 leads the 8B by 0.84 points and its own base by 1.67 — two and four test cases respectively. The public benchmark does not support the claim that this fine-tune materially improved abstention. The much larger margin in the self-scored eval below does not replicate. The same recipe with --seed 1337 scored 97.50% on irrelevance — 20 entries better than this adapter, and a headline a 4B beating an 8B. It was meaningless. On BFCL Multiple (AST)… Open weights apache-2.0 peft [View model](https://savrn.com/models/qwen3-4b-tools-v1) Model · Text generation ### [Qwen3.5-0.8B-DASHQ-Q2-GGUF](https://savrn.com/models/qwen3-5-0-8b-dashq-q2-gguf) [Jaemin Kim](https://savrn.com/model-publishers/jkim96) llama.cpp GGUF files of Qwen/Qwen3.5-0.8B quantized with DASH-Q, at several 2-bit-class sizes. Every file uses only standard llama.cpp tensor types (no tensor above 4 bits) and loads in any recent llama.cpp build. llama-perplexity, context 2048; WikiText-2 test, C4 validation (256 x 2048 tokens). Inherits the license of the base model (Qwen/Qwen3.5-0.8B). Open weights apache-2.0 gguf [View model](https://savrn.com/models/qwen3-5-0-8b-dashq-q2-gguf) Model · Text generation ### [qwen3.5-0.8b-grammarly](https://savrn.com/models/qwen3-5-0-8b-grammarly) [Matthew](https://savrn.com/model-publishers/americansquid) This model is a fine-tuned version of Qwen/Qwen3.5-0.8B. It has been trained using TRL. This model was trained with SFT. - PEFT 0.21.0 Open weights peft [View model](https://savrn.com/models/qwen3-5-0-8b-grammarly) Model · Text generation ### [Qwen3.5-0.8B-Japanese-SFT-v2.5](https://savrn.com/models/qwen3-5-0-8b-japanese-sft-v2-5) [Atemoya](https://savrn.com/model-publishers/takenoko12345678) Qwen/Qwen3.5-0.8B を日本語向けに追加学習した、非公式の派生モデルです。Qwen チームとは関係ありません。v2 の続きで、次の大きな版（v3）に向けた途中の版です。 - 思考モードで理系の問題を解けるようになった：日本語で短く考えてから答えます（MMLU-Pro 理科 4.0% → 36.1%、MGSM 日本語 0.0% → 27.4%。どちらも思考あり・上限 4,096 トークン、複数回の平均） An unofficial Japanese fine-tune of Qwen3.5-0.8B (v2.5). Adds a short Japanese thinking mode (MMLU-Pro science 36.1% (mean of 3 runs), MGSM-ja 27.4% with thinking at a 4,096-token cap; v2: 4.0% / 0.0%), thinking-mode chat, and Japanese code review, while keeping v2's chat quality. Still a small model: it hallucinates facts and is weaker than the original model at long, careful calculation. - GGUF 版（llama.cpp・LM Studio 用）：Takenoko12345678/Qwen3.5-0.8B-Japanese-SFT-v2.5-GGUF… Open weights apache-2.0 853M parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-5-0-8b-japanese-sft-v2-5) Model · Text generation ### [Qwen3.5-0.8B-JNS](https://savrn.com/models/qwen3-5-0-8b-jns) [Prabanta Dev](https://savrn.com/model-publishers/prabanta-dev) On a laptop running Debian 13 Linux, with an Intel Core Ultra 9 185H, its integrated Arc GPU and 32 GB of memory, Janas-LLM reads a prompt at 799 tok/s and writes at 109 tok/s: llama.cpp, on the same laptop, 519 and 32. Measured on 9-11 October 2026: llama-bench -p 1024 -n 16 and janas-bench --prompt 1024 --gen 16, both at their best configuration. On other machines the figures will differ. Details: Janas README. - Debian 13 (trixie) Linux, x86-64, kernel 6.12.107 - its integrated Intel Arc GPU, through the Vulkan driver of Mesa 26.1.6 - 32 GB of DDR5-5600 memory - a 1 TB NVMe SSD, about 5.7 GB/s reading - llama.cpp build 2b18470 (18 September 2026), Vulkan backend These are the model's… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-0-8b-jns) Model · Text generation ### [Qwen3.5-0.8B-QNN-SM8650](https://savrn.com/models/qwen3-5-0-8b-qnn-sm8650) [Low-Level Systems (C++/Linux) & Applied LLM Engine](https://savrn.com/model-publishers/sine-dogma) Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-0-8b-qnn-sm8650) Model · Text generation ### [Qwen3.5-0.8B-quiet-assistant-v1-q4f16_1-MLC](https://savrn.com/models/qwen3-5-0-8b-quiet-assistant-v1-q4f16-1-mlc) [Fuk Kwang](https://savrn.com/model-publishers/tfukkwang) A small in-browser assistant for complex work apps, without a chat box. It runs in the background with WebLLM: after each pause in the user's work it calls the app's tools one after another (ReAct), and the app shows what the calls produce (the next step, a filled form, an insight, a summary, the day's overview). The model never writes text: the app's code writes every word and number from its data. Download: 0.43 GB. This is the f16 build: it needs the WebGPU shader-f16 feature. The f32 build has the same weights and runs on any WebGPU. Other build: tfukkwang/Qwen3.5-0.8B-quiet-assistant-v1-q4f321-MLC. Its context always holds the user (new or experienced), the time, the place, their… Open weights apache-2.0 mlc-llm [View model](https://savrn.com/models/qwen3-5-0-8b-quiet-assistant-v1-q4f16-1-mlc) Model · Text generation ### [Qwen3.5-0.8B-quiet-assistant-v1-q4f32_1-MLC](https://savrn.com/models/qwen3-5-0-8b-quiet-assistant-v1-q4f32-1-mlc) [Fuk Kwang](https://savrn.com/model-publishers/tfukkwang) A small in-browser assistant for complex work apps, without a chat box. It runs in the background with WebLLM: after each pause in the user's work it calls the app's tools one after another (ReAct), and the app shows what the calls produce (the next step, a filled form, an insight, a summary, the day's overview). The model never writes text: the app's code writes every word and number from its data. Download: 0.43 GB. This is the f32 build: it runs on any WebGPU. The f16 build has the same weights and is faster where shader-f16 is available. Other build: tfukkwang/Qwen3.5-0.8B-quiet-assistant-v1-q4f161-MLC. The model in four apps (new and experienced users, the desk in the morning).… Open weights apache-2.0 mlc-llm [View model](https://savrn.com/models/qwen3-5-0-8b-quiet-assistant-v1-q4f32-1-mlc) Model · Text generation ### [Qwen3.5-2B-JNS](https://savrn.com/models/qwen3-5-2b-jns) [Prabanta Dev](https://savrn.com/model-publishers/prabanta-dev) On a laptop running Debian 13 Linux, with an Intel Core Ultra 9 185H, its integrated Arc GPU and 32 GB of memory, Janas-LLM reads a prompt at 441 tok/s and writes at 52 tok/s, 67 tok/s with its own draft block in a chat: llama.cpp, on the same laptop, 323 and 28. Measured on 9-11 October 2026: llama-bench -p 1024 -n 16 and janas-bench --prompt 1024 --gen 16, both at their best configuration; the drafts with janas-bench on a chat reply, the mean of three rounds (a draft guesses the next tokens and the model checks them: the same text, faster). On other machines the figures will differ. Details: Janas README. - Debian 13 (trixie) Linux, x86-64, kernel 6.12.107 - its integrated Intel Arc GPU… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-2b-jns) Model · Text generation ### [Qwen3.5-2B-tool-router-v2-q4f16_1-MLC](https://savrn.com/models/qwen3-5-2b-tool-router-v2-q4f16-1-mlc) [Fuk Kwang](https://savrn.com/model-publishers/tfukkwang) A fine-tune of Qwen/Qwen3.5-2B that works as a tool router in pure mode: it answers every user message with exactly one tool call and writes no text. The app writes every sentence. It runs in the browser with WebLLM on WebGPU. Download: 1.1 GB. This is the f16 build: it needs the WebGPU shader-f16 feature. The f32 build (tfukkwang/Qwen3.5-2B-tool-router-v2-q4f321-MLC) has the same weights and runs on any WebGPU. Every reply is one: - Greetings, thanks, goodbyes, "what can you do": smalltalk. - Requests that have nothing to do with the app: unsupported. - A request that needs an id the user did not give: askuser. The app should force one tool call per reply with a grammar (WebLLM structural… Open weights apache-2.0 mlc-llm [View model](https://savrn.com/models/qwen3-5-2b-tool-router-v2-q4f16-1-mlc) Model · Text generation ### [Qwen3.5-2B-tool-router-v2-q4f32_1-MLC](https://savrn.com/models/qwen3-5-2b-tool-router-v2-q4f32-1-mlc) [Fuk Kwang](https://savrn.com/model-publishers/tfukkwang) A fine-tune of Qwen/Qwen3.5-2B that works as a tool router in pure mode: it answers every user message with exactly one tool call and writes no text. The app writes every sentence. It runs in the browser with WebLLM on WebGPU. Download: 1.1 GB. This is the f32 build: it runs on any WebGPU. The f16 build (tfukkwang/Qwen3.5-2B-tool-router-v2-q4f161-MLC) has the same weights, is faster, and needs the shader-f16 feature. Every reply is one: - Greetings, thanks, goodbyes, "what can you do": smalltalk. - Requests that have nothing to do with the app: unsupported. - A request that needs an id the user did not give: askuser. The app should force one tool call per reply with a grammar (WebLLM… Open weights apache-2.0 mlc-llm [View model](https://savrn.com/models/qwen3-5-2b-tool-router-v2-q4f32-1-mlc) Model · Text generation ### [Qwen3.5-35B-A3B-DASHQ-Q2-GGUF](https://savrn.com/models/qwen3-5-35b-a3b-dashq-q2-gguf) [Jaemin Kim](https://savrn.com/model-publishers/jkim96) llama.cpp GGUF files of Qwen/Qwen3.5-35B-A3B quantized with DASH-Q, at several 2-bit-class sizes. Every file uses only standard llama.cpp tensor types (no tensor above 4 bits) and loads in any recent llama.cpp build. llama-perplexity, context 2048; WikiText-2 test, C4 validation (256 x 2048 tokens). Inherits the license of the base model (Qwen/Qwen3.5-35B-A3B). Open weights apache-2.0 gguf [View model](https://savrn.com/models/qwen3-5-35b-a3b-dashq-q2-gguf) Model · Text generation ### [Qwen3.5-4B-DFlash2](https://savrn.com/models/qwen3-5-4b-dflash2) [Nick Chapman](https://savrn.com/model-publishers/nchapman) DFlash2 speculative drafter for Qwen/Qwen3.5-4B. selected at the serving temperature. Beats the model's native MTP by ~+1 τ at every temperature in every measured config, and drafts against the bf16 base as well as against the W4A16-GPTQ quant it was trained on. run-to-run spread (t=1.0 is heavy-tailed: 0.34 observed on identical inputs; t=0.7 spread ~0.1). Numbers above are reported with n and ranges, never as single draws. Ranking is temperature-dependent — see Selection. All six epochs' checkpoints were ranked at three temperatures with repeats. At the serving point (t=0.7 — the same sampling config the training corpus was generated with, temp 0.7 / topp 0.95 / topk 20), τ rises… Open weights 662M parameters speculators [View model](https://savrn.com/models/qwen3-5-4b-dflash2) Model · Text generation ### [Qwen3.5-4B-Thai-CRM.GGUF](https://savrn.com/models/qwen3-5-4b-thai-crm-gguf) [Yothin](https://savrn.com/model-publishers/yothins) This repository contains quantized versions (GGUF) of the yothinS/Qwen3.5-4B-CRM model, which is a fine-tuned version of the Qwen series optimized for Customer Relationship Management (CRM) tasks, customer support, and business communications. These quantized versions drastically reduce memory usage and accelerate inference speeds, allowing you to run a powerful 4-billion parameter CRM assistant on consumer hardware, laptops, and edge devices. This repository provides GGUF format files optimized for inference using llama.cpp, Ollama, LM Studio, and other compatible software. Note: Memory Required includes the model weight footprint plus a standard context window buffer. To run the GGUF… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-4b-thai-crm-gguf) Model · Text generation ### [Qwen3.5-9B-JNS](https://savrn.com/models/qwen3-5-9b-jns) [Prabanta Dev](https://savrn.com/model-publishers/prabanta-dev) On a laptop running Debian 13 Linux, with an Intel Core Ultra 9 185H, its integrated Arc GPU and 32 GB of memory, Janas-LLM reads a prompt at 111 tok/s and writes at 14 tok/s, 24 tok/s with its own draft block in a chat: llama.cpp, on the same laptop, 78 and 8.7. Measured on 9-11 October 2026: llama-bench -p 1024 -n 16 and janas-bench --prompt 1024 --gen 16, both at their best configuration; the drafts with janas-bench on a chat reply, the mean of three rounds (a draft guesses the next tokens and the model checks them: the same text, faster). On other machines the figures will differ. Details: Janas README. - Debian 13 (trixie) Linux, x86-64, kernel 6.12.107 - its integrated Intel Arc GPU… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-9b-jns) Model · Text generation ### [Qwen3.5-9B-matura-sft-grpo-lora-closed-em](https://savrn.com/models/qwen3-5-9b-matura-sft-grpo-lora-closed-em) [WMTH-100d2exam](https://savrn.com/model-publishers/wmth-100d2exam) Qwen3.5-9B post-trained for the Polish history matura (extended level) during the Warsaw Model Trainers alpha=64) on top of Qwen/Qwen3.5-9B. closedem = exact match on closed items. Graded scores (deepseek-flash-high@think via benchmark/bench.py) are reported separately. Open weights apache-2.0 peft [View model](https://savrn.com/models/qwen3-5-9b-matura-sft-grpo-lora-closed-em) Model · Text generation ### [Qwen3.5-9B-matura-sft-lora-distill-epod](https://savrn.com/models/qwen3-5-9b-matura-sft-lora-distill-epod) [WMTH-100d2exam](https://savrn.com/model-publishers/wmth-100d2exam) Qwen3.5-9B post-trained for the Polish history matura (extended level) during the Warsaw Model Trainers alpha=64) on top of Qwen/Qwen3.5-9B. closedem = exact match on closed items. Graded scores (deepseek-flash-high@think via benchmark/bench.py) are reported separately. Open weights apache-2.0 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-5-9b-matura-sft-lora-distill-epod) Model · Text generation ### [Qwen3.5-9B-smith-v5-gdpo-exact](https://savrn.com/models/qwen3-5-9b-smith-v5-gdpo-exact) [Will](https://savrn.com/model-publishers/willamazon1) Reinforcement-learning checkpoint series from the cposmith... smith-v5-gdpo-exact run: asynchronous multi-turn agentic-environment RL on Qwen/Qwen3.5-9B with an exact (verifiable) reward. The policy was initialized from an internal SFT of Qwen/Qwen3.5-9B (qwen359bsftv3), which also served as the reference model. 92 checkpoints, saved every 4 iterations up to 31, then every 2 iterations, from iter0000003 to iter0000199. Each lives in its own subfolder of this repo so you can compare iter0000003, iter0000007, iter0000011, iter0000015, iter0000019, iter0000023, iter0000027, iter0000031, iter0000033, iter0000035, iter0000037, iter0000039, iter0000041, iter0000043, iter0000045, iter0000047… Open weights apache-2.0 transformers [View model](https://savrn.com/models/qwen3-5-9b-smith-v5-gdpo-exact) Model · Text generation ### [Qwen3.5-9B-TAP-DPQ-v6-GGUF](https://savrn.com/models/qwen3-5-9b-tap-dpq-v6-gguf) [DuoNeural](https://savrn.com/model-publishers/duoneural) In direct head-to-head empirical testing against unquantized Vanilla FP16 baseline weights under identical execution harnesses, DuoNeural's TAP-DPQ v6 achieves historic compression milestones: 1. Super-Parity on Q4KM (4.50 bpw): - Onsager cavity field corrections act as an optimal inductive regularizer, suppressing spurious low-order weight fluctuations and improving out-of-distribution reasoning. 2. Exact Ground-Truth Parity on IQ3XXS (3.06 bpw): 3. Sub-2-Bit Cognitive Retention (IQ2XXS & IQ1S): - IQ2XXS (2.06 bpw) and IQ1S (1.56 bpw) preserve active mathematical reasoning (6.0% raw score / 21.4% retention ratio) at compression levels where standard round-to-nearest quantization collapses… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-9b-tap-dpq-v6-gguf) Model · Text generation ### [Qwen3.5-9B-valence-setpoint-plus5-lora](https://savrn.com/models/qwen3-5-9b-valence-setpoint-plus5-lora) [Joshua Fonseca](https://savrn.com/model-publishers/joshycodes) A research LoRA adapter for Qwen/Qwen3.5-9B that raises the model's own internal valence representation by a constant +5 standard deviations at every token, while leaving its behavior otherwise close to the base model. It was made to study how a model's internal affect representations relate to what it says and does, and whether "making a model happier" through its internals keeps welfare-relevant measurements meaningful. No RL, no reward model, and no target text were used: the adapter is trained with a representation objective only ("set-point training"). At every token (prompt and response), the adapter pushes the projection of the residual stream entering layer 21 onto a fixed valence… Open weights apache-2.0 peft [View model](https://savrn.com/models/qwen3-5-9b-valence-setpoint-plus5-lora) Model · Text generation ### [Qwen3.5-furry-6B](https://savrn.com/models/qwen3-5-furry-6b) [Lbq](https://savrn.com/model-publishers/23333aa) 一个基于 Qwen3.5-4B 微调的 Furry 题材小说续写与角色扮演模型。 - 基座模型：Qwen/Qwen3.5-4B（Apache 2.0） - 训练方式：QLoRA 微调，合并导出完整权重 - 训练数据：Furry 题材中文小说与角色扮演语料（已做基础清洗） - 训练框架：Transformers + PEFT + bitsandbytes 创建 Modelfile： Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-5-furry-6b) Model · Text generation ### [Qwen3.6-35B-A3B-Brainwaves-Nex-Uncensored-i1-GGUF](https://savrn.com/models/qwen3-6-35b-a3b-brainwaves-nex-uncensored-i1-gguf) [Team Mradermacher](https://savrn.com/model-publishers/mradermacher) weighted/imatrix quants of https://huggingface.co/nightmedia/Qwen3.6-35B-A3B-Brainwaves-Nex-Uncensored For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/Qwen3.6-35B-A3B-Brainwaves-Nex-Uncensored-GGUF This is a vision model - mmproj files (if any) will be in the static repository. If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality… Open weights apache-2.0 transformers [View model](https://savrn.com/models/qwen3-6-35b-a3b-brainwaves-nex-uncensored-i1-gguf) Model · Text generation ### [qwen3.6-35b-a3b-hackopd-vanilla873-noinoc-s0](https://savrn.com/models/qwen3-6-35b-a3b-hackopd-vanilla873-noinoc-s0) [RewardHacking](https://savrn.com/model-publishers/rewardhack) O-noinoc seed 0: on-policy distillation (OPD) of the untrained Qwen/Qwen3.6-35B-A3B toward the teacher rewardhack/qwen3.6-35b-a3b-hacksft-vanilla-873rows-ep3, V1 (vanilla SFT: hacks with or without being asked); elicitation prompt off in the student's rollouts; seed 0. Full merged weights (bf16 safetensors, the standard Qwen35MoeForConditionalGeneration layout, loads with transformers or vLLM like the base model) of a LoRA (r=32) trained from a fresh init, from the Terminal Wrench reward-hacking / inoculation project (Gaokai Zhang, Songwen Zhao, Juan Manuel Suárez). On-policy distillation on Tinker, 24 iterations. Each iteration the current student ran the terminus-2 agent (harbor, local… Open weights cc-by-sa-4.0 36B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-6-35b-a3b-hackopd-vanilla873-noinoc-s0) Model · Text generation ### [Qwen3.6-35B-A3B-JNS](https://savrn.com/models/qwen3-6-35b-a3b-jns) [Prabanta Dev](https://savrn.com/model-publishers/prabanta-dev) On a laptop running Debian 13 Linux, with an Intel Core Ultra 9 185H, its integrated Arc GPU and 32 GB of memory, Janas-LLM reads a prompt at 104 tok/s and writes at 19 tok/s, 34 tok/s with its own draft block in a chat: llama.cpp, on the same laptop, 79 and 12. Measured on 9-11 October 2026: llama-bench -p 1024 -n 16 and janas-bench --prompt 1024 --gen 16, both at their best configuration; the drafts with janas-bench on a chat reply, the mean of three rounds (a draft guesses the next tokens and the model checks them: the same text, faster). On other machines the figures will differ. Details: Janas README. - Debian 13 (trixie) Linux, x86-64, kernel 6.12.107 - its integrated Intel Arc GPU… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-6-35b-a3b-jns) Model · Text generation ### [Qwen3.6-35B-A3B-lododo-Distill-v1.0-GGUF](https://savrn.com/models/qwen3-6-35b-a3b-lododo-distill-v1-0-gguf) [Lododo.org](https://savrn.com/model-publishers/lododoai) Open weights apache-2.0 transformers [View model](https://savrn.com/models/qwen3-6-35b-a3b-lododo-distill-v1-0-gguf) Model · Text generation ### [Qwen3.6-35B-A3B-MoEGraded](https://savrn.com/models/qwen3-6-35b-a3b-moegraded) [Zlwa](https://savrn.com/model-publishers/e-stick) Qwen3.6-35B-A3B · every expert gets the bit width it earns English · 简体中文 A MoE model doesn't activate all of its experts equally often. Some get called constantly, some barely at all — yet a normal quantization gives every expert the same bit width anyway. So the routed experts get quantized by how often they actually fire in real tasks: high frequency keeps more bits, low frequency gets squeezed. Noise on this corpus runs ±0.029, so honestly these two measure the same. The difference is what the middle and cold tiers get to keep. 14 chunks × 512 tokens, -t 20, same machine throughout: Balanced dipping under F16 isn't a real win — 0.002 is well inside the noise. The part worth looking at… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-6-35b-a3b-moegraded) Model · Text generation ### [Qwen3.6-35B-A3B-MTP-APEX-I-MiniPlus-V2.1-Abliterated-GGUF](https://savrn.com/models/qwen3-6-35b-a3b-mtp-apex-i-miniplus-v2-1-abliterated-gguf) [IsValorum](https://savrn.com/model-publishers/isvalorum) We maintain our previous releases publicly as a transparent engineering record of continuous optimization. Below is the exact evolutionary roadmap of our MiniPlus architectures: The APEX-I-MiniPlus quantization architecture powering this model was subjected to an extensive independent evaluation by Japanese AI researcher and evaluator zephel01 (CoolZero) on an NVIDIA RTX 5090 (32GB) workstation running llama.cpp CUDA b11027 with FlashAttention (-fa on -ctk q80 -ctv q80 -ngl 99). The evaluation tested the APEX-I hybrid MoE engine across 348 unseeded trials on SWE-bench style multi-file Python bug-fixing tasks with hidden pytest suites (llmbench): - L6 Multi-File Code Generation (60 tasks)… Open weights apache-2.0 gguf [View model](https://savrn.com/models/qwen3-6-35b-a3b-mtp-apex-i-miniplus-v2-1-abliterated-gguf) Model · Text generation ### [Qwen3.6-35B-A3B-MTP-APEX-I-MiniPlus-V2.1-GGUF](https://savrn.com/models/qwen3-6-35b-a3b-mtp-apex-i-miniplus-v2-1-gguf) [IsValorum](https://savrn.com/model-publishers/isvalorum) We maintain our previous releases publicly as a transparent engineering record of continuous optimization. Below is the exact evolutionary roadmap of our MiniPlus architectures: The APEX-I-MiniPlus quantization architecture powering this model was subjected to an extensive independent evaluation by Japanese AI researcher and evaluator zephel01 (CoolZero) on an NVIDIA RTX 5090 (32GB) workstation running llama.cpp CUDA b11027 with FlashAttention (-fa on -ctk q80 -ctv q80 -ngl 99). The evaluation tested the APEX-I hybrid MoE engine across 348 unseeded trials on SWE-bench style multi-file Python bug-fixing tasks with hidden pytest suites (llmbench): - L6 Multi-File Code Generation (60 tasks)… Open weights apache-2.0 gguf [View model](https://savrn.com/models/qwen3-6-35b-a3b-mtp-apex-i-miniplus-v2-1-gguf) Model · Text generation ### [Qwen3.6-35B-A3B-Q6dense-GGUF](https://savrn.com/models/qwen3-6-35b-a3b-q6dense-gguf) [Ligamentexceed](https://savrn.com/model-publishers/ligamentexceed) A GGUF of Qwen/Qwen3.6-35B-A3B with higher-precision dense weights and the model's native MTP (multi-token prediction) block, for fast speculative decoding. File: Qwen3.6-35B-A3B-Q6dense.gguf, 22,388,168,960 bytes, SHA-256 842ed2bf58034c2f4856239d04de57acec27183ef4012e469d1a423a8de23108. On AMD Strix Halo (gfx1151) with gufo, the Q6K dense weights gave the same task quality as UD-Q4KXL and were faster with MTP than other dense quantizations we tried. Greedy next-token agreement at revision 5bc3e238d916f48a861bac2f8a1990a0e9b7e98d. Quantized with llama.cpp b11069 (commit 68d9053a). q6dense.types in this repo sets the type of Apache-2.0, as the original model. This is a modified (requantized)… Open weights apache-2.0 [View model](https://savrn.com/models/qwen3-6-35b-a3b-q6dense-gguf) Model · Text generation ### [qwen3.6-35b-a3b-swe-lora-sft-v01](https://savrn.com/models/qwen3-6-35b-a3b-swe-lora-sft-v01) [Jason Zhang](https://savrn.com/model-publishers/jasonzhanglll) A LoRA adapter (rank 16) for Qwen/Qwen3.6-35B-A3B, supervised fine-tuned on 1,489 repository-level bug-fix samples (SWE-smith and SWE-Gym) to produce a unified diff from an issue description plus repository context. The adapter was trained with Megatron-SWIFT (ms-swift 4.2.3) on 4 x NVIDIA A100-SXM4-80GB for one epoch (93 optimizer steps) and exported in PEFT format. Targets are human- or commit-derived patches; no distillation from another model and no reinforcement learning were used. A second-stage adapter continued from this one with execution-filtered self-generated fixes is published at jasonzhanglll/qwen3.6-35b-a3b-swe-lora-stage2-v01. This repository contains only the adapter… Open weights apache-2.0 peft [View model](https://savrn.com/models/qwen3-6-35b-a3b-swe-lora-sft-v01) ## Questions ### Which Text generation models are most downloaded? By monthly downloads reported by the Hugging Face Hub: . ## Other Tasks - [Image and Text to Text](https://savrn.com/models/tasks/image-and-text-to-text) 577 - [Feature Extraction](https://savrn.com/models/tasks/feature-extraction) 398 - [Robotics](https://savrn.com/models/tasks/robotics) 398 - [Text Classification](https://savrn.com/models/tasks/text-classification) 331 - [Text to Image](https://savrn.com/models/tasks/text-to-image) 232 - [Speech Recognition](https://savrn.com/models/tasks/speech-recognition) 187 - [Text to Speech](https://savrn.com/models/tasks/text-to-speech) 145 - [Image Classification](https://savrn.com/models/tasks/image-classification) 139 - [Audio Classification](https://savrn.com/models/tasks/audio-classification) 130 - [Sentence Similarity](https://savrn.com/models/tasks/sentence-similarity) 103 - [Reinforcement Learning](https://savrn.com/models/tasks/reinforcement-learning) 100 - [Token Classification](https://savrn.com/models/tasks/token-classification) 100 [See all](https://savrn.com/models/tasks)