SAVRN Model Hub · Models by Task
Text Generation Models
1,582 open-weight text generation models in the SAVRN Model Hub, with Convergent Intelligence, Qwen and Francesca Padovani publishing the most.
1,582Models
732Publishers
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.
Most Downloaded
| Model | Publisher | Parameters | License | Monthly downloads | Cheapest GPUs at 16-bit |
|---|---|---|---|---|---|
| Qwen3-0.6B | Qwen | 752M | apache-2.0 | 25.2M | 1x MI300X, $1.85/hr |
| gpt2 | OpenAI community | 137M | mit | 18.8M | 1x MI300X, $1.85/hr |
| Qwen3-4B | Qwen | 4B | apache-2.0 | 11.4M | 1x MI300X, $1.85/hr |
| Qwen3-14B | Qwen | 14.8B | apache-2.0 | 9.4M | 1x MI300X, $1.85/hr |
| Qwen3-8B | Qwen | 8.2B | apache-2.0 | 8.1M | 1x MI300X, $1.85/hr |
| Qwen2.5-7B-Instruct | Qwen | 7.6B | apache-2.0 | 6.4M | 1x MI300X, $1.85/hr |
| gpt-oss-20b | OpenAI | 20.9B | apache-2.0 | 5.7M | 1x MI300X, $1.85/hr |
| opt-125m | AI at Meta | — | other | 5.6M | — |
| DeepSeek-V3.2 | DeepSeek | 685.4B | mit | 5.4M | 7x MI325X, $14.00/hr |
| Qwen2.5-0.5B-Instruct | Qwen | 494M | apache-2.0 | 5.3M | 1x MI300X, $1.85/hr |
Licenses
| License | Models | Commercial use |
|---|---|---|
| apache-2.0 | 734 | Yes |
| not stated | 345 | Not stated |
| other | 175 | Read the license |
| mit | 152 | Yes |
| cc-by-nc-4.0 | 42 | Not without separate permission |
| llama3.2 | 33 | Read the license |
| gemma | 22 | Yes, with conditions |
| llama3.1 | 21 | Yes, with conditions |
Who Publishes Them
| Publisher | Models |
|---|---|
| Convergent Intelligence | 54 |
| Qwen | 42 |
| Francesca Padovani | 37 |
| Yanggangu | 28 |
| Shreyansh singh | 25 |
| Kumar | 23 |
All 1,582 Models, Page 13 of 27
The EXL3 quantization made by Mia-AiLab of deepseek-ai/DeepSeek-V4.1-Flash, repackaged as GGUF without re-quantization. Requires ranma.cpp ranma20261013 or newer (github.com/saturnsky/ranma.cpp).
dealignai/DeepSeek-V4.1-Flash-UNCENSORED-EXL3-2.9bpw (abliteration and EXL3 quantization by dealignai), repackaged as GGUF without re-quantization. Requires ranma.cpp ranma20261013 or newer (github.com/saturnsky/ranma.cpp).
DeepSeek-V4-Pro-0813 is the official release of DeepSeek-V4-Pro, superseding the preview version, with greatly enhanced agentic capabilities and performance improvements that are especially pronounced in production environments. It is built on the DeepSeek-V4-Pro (Preview) model structure, with a DSpark speculative decoding module attached. DeepSeek-V4-Pro-0813 outperforms DeepSeek-V4-Pro (Preview) on the benchmarks listed below, and is broadly competitive with the strongest proprietary models available. 1. For the code-agent tasks among the public benchmarks above, DeepSeek-V4-Pro-0813 is evaluated with the minimal mode of DeepSeek Harness as the agent framework, using the max reasoning…
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Merged checkpoint produced by the family-aware Delta-P2S experiment package.
HorneSci's open-weight continued-pretraining release for Qwen3.8-27B. DeltaRoutingAI trains language adapters on Python technical documentation and delivers a 4.61% reduction in held-out text NLL in its paired evaluation. The downloadable checkpoint is a PEFT LoRA adapter in safetensors format (233.6 MB). Load it with Qwen/Qwen3.8-27B, pinned to revision 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0. Qwen authors the base model; HorneSci authors the continued-pretraining recipe and adapter weights. The adapter weights are released under Apache 2.0 with upstream attribution and data notices included. Use a CUDA environment with the versions in requirements-inference.txt. The evaluated environment…
Diakosmos (διάκοσμος, "the ordering of a world") is a byte-level harmonic recurrent language model, 99,823,934 parameters, written and trained from scratch in the Alpha language on a rented RTX 3070. It has no tokenizer and no attention matrix: each byte enters as a fixed constellation of phasors (the wiretable), and each of 14 blocks blends a decaying harmonic recurrence over the past with a planar rotation between seven wire groups. Its design is harmonic-gpt's leading model. The tensors carry harmonic-gpt's parameter names; wiretable is the fixed byte encoding (256 x 1330). updates of 64 windows, 24.5 GPU hours and 5.29 kWh on one RTX 3070. Held-out OpenWebText: 1.4194 bits per byte…
A LoRA adapter for Qwen/Qwen2.5-0.5B-Instruct that turns a short user profile (age, gender, height, weight, activity level, dietary preference, daily calorie target) into a one-day meal plan with breakfast, lunch, snack and dinner. It is a small, fast, single-purpose model: about 2.2M trainable parameters on top of a 0.5B base, usable on CPU. The adapter was trained on one fixed prompt layout. Use the same system prompt and the same field names and order, otherwise quality drops. Two details matter: - Pass the system prompt explicitly. Without it, the chat template inserts Qwen's default system prompt, which the adapter never saw. - The first field is spelled Ages: in the training data, so…
Diplodocus preview 4, the Osaurus assistant for finance, documents, spreadsheets and tool-assisted work, built on IFM/K2-Horizon-7B and trained on Osaurus' own workspace tools (workbooks and documents edited in place, the built-in calculate tool, reasoning on and off). This is the GGUF build for llama.cpp: one calibrated Q6K file, 7.76 GB. Using Osaurus on a Mac? Use the MLX bundle OsaurusAI/Diplodocus-preview-JANG6M instead. Same weights, same template, same reasoning toggle. Requires llama.cpp with K2 Horizon support (ggml-org/llama.cpp#29535, merged 2026-10-06). Older builds, and apps that bundle an older llama.cpp, will not load the k2-horizon architecture. Trained rounds R8–R11 on top…
This model is a fine-tuned version of cosmos1030/gmp-kd3e-1-s80pct-lr1e-420260916220740 on the trl-lib/ultrafeedbackbinarized dataset. It has been trained using TRL. This model was trained with DPO, a method introduced in Direct Preference Optimization: Your Language Model is Secretly a Reward Model.
DriveTwinLLM is an instruction-tuned, physics-grounded automotive diagnostic language model built on top of Qwen/Qwen2.5-3B-Instruct. It was fine-tuned using LoRA on NVIDIA B200 GPU hardware (180 GB VRAM) to achieve 98.13% evaluation token accuracy on real-world vehicle trouble codes, live ECU telemetry, and OEM service manuals. The model enforces strict zero-hallucination guardrails and features a specialized Dual-Persona Architecture: - Diagnostic computer scan fee ($90 – $130) - Certified shop labor ($120/hr flat-rate standard) - Strictly refuses unmonitored sensors (transmission fluid temperature, TPMS, cabin humidity) and invalid codes (P9999), redirecting the user to monitored engine…
本地改写页面 Drop AIGC： https://github.com/k856487/Drop-AIGC Dailv1.5-f16.gguf 为 F16 GGUF，约 3.1 GB，由 llama.cpp 加载。 页面的环境、启动方式与可选的推理服务地址，见 GitHub 仓库说明。
libraryname: transformers - neuromorphic - connectomics - drosophila - flywire - sparse-computation - causal-lm - qwen - flywire-codex-2024 pipelinetag: text-generation Biological Substrate: Adult Drosophila melanogaster Whole-Brain Connectome (FlyWire / Codex, Nature 2024) Contemporary Large Language Models rely universally on dense, rectangular projections ($O(N^2)$). Every token representation is forced through uniform, all-to-all matrix transformations that treat every latent channel with identical geometric indifference. Biological nervous systems operate under radically different evolutionary constraints. The adult fruit fly (Drosophila melanogaster) coordinates complex 3D aerobatics…
Official weights for the Dual-Loop Cognitive Controller on Qwen/Qwen3.5-2B ($D=2048$, Layer 11 hook, ~110M parameter deliberation adapter). The Dual-Loop Controller provides hardware-aligned, non-autoregressive System 2 deliberation directly within the latent residual stream of modern language models. It enables models to recursively deliberate in continuous hidden space without generating costly Chain-of-Thought (CoT) text tokens, eliminating KV-cache explosion and 30–60 second generation latencies. This adapter is strictly designed, calibrated, and hooked into the architectural dimensions of Qwen/Qwen3.5-2B: 1. Cognitive Matrix Helper (Tversky Elimination-by-Aspects): - Evaluates options…
Roy C Abstract. We present a four-stage pipeline for producing small language models (1.7B parameters) that exhibit self-critiquing dual-cognition reasoning from a 30B-parameter teacher. The pipeline chains: (1) proof-weighted knowledge distillation from three teacher variants (Instruct, Thinking, Coder) of Qwen3-30B-A3B, each producing distinct capability profiles in the student; (2) Topological Knowledge Distillation (TKD), which uses the Discrepancy Calculus (DISC) framework to decompose the teacher's output distribution into smooth, jump, and singular-continuous components via bounded variation theory, allocating training capacity to structural boundaries that standard KD smears across…
从零训练的 38.85M 参数英文小模型，消费级显卡（RTX 4060 Laptop 8GB）单卡训完。 本仓库同时提供了 GGUF 格式（由 model/litmodel.pth 直接转换，模板保持与训练一致），可直接用 Ollama 导入： scripts/chat.sh 里的三种问法： 在 Python 里直接用： - 只会做被交代的任务：给它一句裸问题（如 What is a volcano?）它不会回答；必须像上面那样把任务写清楚。 - 上下文上限 512 token：prompt + 生成长度之和；超出会报 Cannot forward sequence of length T。 - 预训练语料：TinyStoriesV2-GPT4（CDLA-Sharing-1.0）与 Children-Stories（用户自备，含 GPT 生成内容） - 本包模型权重由上述数据从零训练得到，仅供研究与个人使用；若再分发请一并保留此说明与 MODELCARD.md。
This is a frozen snapshot of CL, EDOS Engineering's experimental continual-learning chatbot, deployed under the persona "Hades." CL learns while it runs by editing the model's original quantized weight tensors in place. Its training is self-directed: the model selects its own training corpus, decides what is important, and is then trained on what it selected. It reads new AI research papers in full, judges which findings matter, writes the training text itself as attributed reviews, and marks which of them to learn and how strongly. It also learns from conversation. It uses no LoRA, adapter or side network. This export is the live model after 1,091 online learning updates, merged into one…
ELF-B, a continuous diffusion language model, trained on the embeddings of T5Gemma-2-270M. From the paper Scaling and Distilling Text Embeddings for Better Diffusibility. ELF (Embedded Language Flow) is a continuous diffusion language model that generates a sequence of text embeddings and decodes them to tokens. This model is the unchanged ELF-B trained on OpenWebText-1024 (sequences of 1024 tokens) for 5 epochs at a batch of 512, on the frozen embeddings of the T5Gemma-2 encoder. Sampling uses the SDE sampler with self-conditioning guidance; --sc sets the guidance scale and --nfe the number of steps. With the code of the paper; the checkpoint and the tokenizer are downloaded from this…
ELF-B, a continuous diffusion language model, trained on the embeddings of T5Gemma-2-270M-OWTdistilled, distilled from T5Gemma-2. From the paper Scaling and Distilling Text Embeddings for Better Diffusibility. ELF (Embedded Language Flow) is a continuous diffusion language model that generates a sequence of text embeddings and decodes them to tokens. This model is the unchanged ELF-B trained on OpenWebText-1024 (sequences of 1024 tokens) for 5 epochs at a batch of 512, on the frozen embeddings of the distilled student T5Gemma-2-270M-OWTdistilled. Sampling uses the SDE sampler with self-conditioning guidance; --sc sets the guidance scale and --nfe the number of steps. With the code of the…
ELF-M, a continuous diffusion language model, trained on the embeddings of T5Gemma-2-270M. From the paper Scaling and Distilling Text Embeddings for Better Diffusibility. ELF (Embedded Language Flow) is a continuous diffusion language model that generates a sequence of text embeddings and decodes them to tokens. This model is the unchanged ELF-M trained on OpenWebText-1024 (sequences of 1024 tokens) for 5 epochs at a batch of 512, on the frozen embeddings of the T5Gemma-2 encoder. Sampling uses the SDE sampler with self-conditioning guidance; --sc sets the guidance scale and --nfe the number of steps. With the code of the paper; the checkpoint and the tokenizer are downloaded from this…
ELF-M, a continuous diffusion language model, trained on the embeddings of T5Gemma-2-270M-OWTdistilled, distilled from T5Gemma-2. From the paper Scaling and Distilling Text Embeddings for Better Diffusibility. ELF (Embedded Language Flow) is a continuous diffusion language model that generates a sequence of text embeddings and decodes them to tokens. This model is the unchanged ELF-M trained on OpenWebText-1024 (sequences of 1024 tokens) for 5 epochs at a batch of 512, on the frozen embeddings of the distilled student T5Gemma-2-270M-OWTdistilled. Sampling uses the SDE sampler with self-conditioning guidance; --sc sets the guidance scale and --nfe the number of steps. With the code of the…
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019). - PEFT 0.20.0
EMENDER E97 4B is a 4,045,972,080-parameter hybrid recurrent language model trained from scratch (~100B tokens), then developed into an observation-driven conversational tool agent through a verified post-training program: task proposals are executed by a real coding-agent harness (Pi), and only deterministically verified trajectories enter training. Every capability claim below was measured on frozen evaluation panels under a fixed protocol, with full audit receipts retained in the release folders. The first checkpoint in the program's history to pass its frozen dual gate. All thresholds passed on a single measurement with no re-rolls: OpenHands-compatible environment, autonomous…
This model is a fine-tuned version of fpadovani/eng-latn-10mb-ppt-Dp-100mb-packedseed3407. It has been trained using TRL. This model was trained with SFT.
This model is a fine-tuned version of fpadovani/eng-latn-10mb-ppt-Dp-10mb-packedseed3407. It has been trained using TRL. This model was trained with SFT.
OpenESM d12 model trained on ClimbMix. This repository contains the pretraining checkpoint without optimizer state conversion. - model.ckpt: Lightning checkpoint. - modelingesm.py: standalone model and checkpoint loader. - tokenizer/tokenizer.pkl: serialized ESM tokenizer. - tokenizer/tokenbytes.pt: token byte table for BPB evaluation. The 160M label describes the model parameter count. Training token counts are intentionally omitted from the repository name and file names. The code and checkpoint format are maintained in the
OpenESM d26 model trained on ClimbMix. This repository contains the pretraining checkpoint without optimizer state conversion. - model.ckpt: Lightning checkpoint. - modelingesm.py: standalone model and checkpoint loader. - tokenizer/tokenizer.pkl: serialized ESM tokenizer. - tokenizer/tokenbytes.pt: token byte table for BPB evaluation. The 1B label describes the model parameter count. Training token counts are intentionally omitted from the repository name and file names. The code and checkpoint format are maintained in the
OpenESM d20 model trained on ClimbMix. This repository contains the pretraining checkpoint without optimizer state conversion. - model.ckpt: Lightning checkpoint. - modelingesm.py: standalone model and checkpoint loader. - tokenizer/tokenizer.pkl: serialized ESM tokenizer. - tokenizer/tokenbytes.pt: token byte table for BPB evaluation. The 520M label describes the model parameter count. Training token counts are intentionally omitted from the repository name and file names. The code and checkpoint format are maintained in the
Quantized INT4 weights designed to run locally on ESP32-S3 with 16MB Flash and Octal PSRAM. This model repository contains the binary weights, metadata, license, and tokenizer assets for the esp32-sandbox project (projects/s3-tiny-stories). - README.md - Model Card and documentation - LICENSE - Apache 2.0 License - metadata.json - Hardware, quantization, and model architecture metadata -.bin - Compiled INT4 model weights - tokenizer.json - SentencePiece / BPE vocabulary configuration Download the binary file (model.bin) and flash it directly to your ESP32-S3: Refer to the esp32-sandbox repository for firmware building, flashing, and serial monitoring.
Model · Text generation
eus-100mb-after-wc-loglinear-newlex-before-ckpt500-packed-bfdiso_seed3407
This model is a fine-tuned version of francesca9805/ppt-wc-loglinear-newlex-eus-before-100mb-packed-bfdisoseed3407. It has been trained using TRL. This model was trained with SFT.
This model is a fine-tuned version of francesca9805/eus-latn-100mb-ppt-mp-struct-100mbseed10. It has been trained using TRL. This model was trained with SFT.
evilguy 3 is the current version of evilguy, a fine-tune of Qwen3 14B with a specific personality: lazy, rude, moody, good at math, fluent in internet slang and memes, and unwilling to make things up. It is not meant to be a helpful assistant. It is meant to be a character. I built it by writing and generating about 7,900 conversation examples and training a LoRA on top of Qwen3 14B using Unsloth. Qwen3 runs with thinking disabled, so replies stay short instead of turning into long reasoning. The result runs locally in llama.cpp as a GGUF file. - evilguy 3 (this one): Qwen3 14B. Thinking disabled, about 7,900 examples, tool calling, slang and meme knowledge, and stronger handling of…
This branch contains the final LoRA adapter for selection seed 42. Training seed is 42 for every branch. main contains seed 42; seed1234 and seed2026 contain the other selection seeds. dataselection/selections/accuracyhard10pctseedsweep/seed42.jsonl in the source project. GRPO prompts use the lgconvfinmcqgrpo formatter, which uses the convfinqagrpo prompt template. 8 generations, reward weights 1.0 for answer correctness and 0.1 for format. before publication. This repository contains adapters, not merged base weights. No validation benchmark was run as part of this training job. The training status and final adapter files were checked before upload. Use…
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Model · Text generation
f-cov-l0-0-no-eos-qwen3-1.7b-compression-bs32-n16-32k-146103-final
Standard BF16 Hugging Face model, exported from training step 100. Weights, configuration, tokenizer, and the thinking chat template are at the repository root. Source checkpoint: hi-todayis-jh/f-cov-l0-0-no-eos-qwen3-1.7b-compression-bs32-n16-32k-146103-step100 at commit a4f98caceb123c954e6a6b176ec2e9b666fc970f. Load this repository directly with AutoModelForCausalLM.frompretrained or vLLM. Use enablethinking=True in the tokenizer chat template for the existing thinking evaluations. The export manifest records source provenance, file checksums, tensor validation, and a successful Transformers CPU load and forward pass.
This repository provides an instruction-tuned causal language model for text generation and chat-style prompts. The configured snapshot destination is available here. Use a current release of transformers to load the model and tokenizer: For long inputs, use context settings supported by the installed runtime and account for available memory. See LICENSE for the applicable terms.
28 layers, width 1536, 16 heads, SwiGLU FFN 4096, RoPE, RMSNorm, tied embeddings, context 256. IMPORTANT: Fresh weights trained from scratch on English FineWeb-Edu (language modeling) and Smol-SmolTalk (chat). This model is experimental. A public checkpoint is not a guarantee of fluent answers. Consult trainingstate.json to check its training phase and token counts. Tokenizer: new ByteLevel BPE with 16,000 IDs. Chat prompts use question and responses end in. Training sources: https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu and https://huggingface.co/datasets/HuggingFaceTB/smol-smoltalk. Respect upstream dataset terms. Training uses a single Kaggle GPU and restarts across sessions.…
Koleksi checkpoint model dan LoRA adapters untuk penalaran hukum pidana Indonesia berbasis putusan peradilan tingkat pertama Mahkamah Agung Republik Indonesia. Proyek riset ini mengevaluasi adaptasi domain peradilan (Supervised Fine-Tuning & QLoRA) dari model berskala ringan (1.5B/3B) hingga skala besar (12B, 27B, dan 31B) untuk menyusun pertimbangan hukum (ratio decidendi) dan amar putusan (dictum) yang setia pada fakta persidangan dan patuh pada asas pembuktian Pasal 184 KUHAP. Repository ini diorganisasikan sebagai satu project induk bertingkat (multi-folder model repository), di mana setiap arsitektur model, checkpoint per-epoch, dan artefak evaluasi dikelompokkan ke dalam subdirektori…
This model is served in https://github.com/xelananv/fhe-ssm: it is a 404,715,520 parameter recurrent language model whose every (!) nonlinearity is a low-degree polynomial with trained coeffs, so that the CKKS encrypted evaluation computes precisely the model's own float64 arithmetic without any post-hoc approximation of softmax or GELU. It succesfully generated text for 64 prompts at once under encryption, autoregressively, for 148 consecutive steps (=> 9,472 tokens) at 128 bit classical parameters, matching its own plaintext top-1 choice 99.7 % of the time. Those are properties of the circuit; this card is about the model, which is small and was trained on a few hundred dollars of…
Meridian.AI is a finance-specialized language model that continuously fine-tunes a Qwen2.5-0.5B backbone every hour on 25+ finance and math datasets, using Elastic Weight Consolidation (EWC) to prevent catastrophic forgetting across training sessions. The entire pipeline runs unattended on free GitHub Actions infrastructure — no GPUs. The deployed checkpoint is a standard Qwen2 model — trustremotecode=True is not required. Inputs are formatted with the ### Instruction: / ### Response: template used during training. A weighted streaming mix of 25+ finance and instruction datasets, including gbharti/finance-alpaca, sujet-ai/Sujet-Finance-Instruct-177k, nvidia/OpenMathInstruct-2…
Using llama.cpp release b11259 for quantization. Don't know which to choose? Grab Q4KM (17.20GB) - usually a good mix of size and performance. Download instructions available here First, make sure you have the Hugging Face CLI installed: The files marked true in the Split column above are stored as multiple parts in a folder. To download all the parts to a local folder, run: You can either specify a new local-dir (FINAL-BenchDarwin-27B-RSI-bf16) or download them all in place (./) These quants run with llama.cpp - installable in one line via llama.app: llama-server includes a built-in chat web UI, served at http://localhost:8080 by default. These quants were made with llama.cpp release…
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019). - PEFT 0.20.0
Fleet is a dynamically extensible Semantic Inference Language Model (SiLM) built on Qwen/Qwen3-4B-Instruct-2507. Fleet is not an external prompt router. The routing system lives inside the causal language model itself through the custom FleetForCausalLM architecture. Fleet replaces selected Qwen projections with native FleetLinear modules. The model also contains a dynamic semantic prototype bank: Routing therefore happens inside model inference, rather than through an external classifier that chooses a different model before generation. This checkpoint currently contains six specialists: - mathreasoning - medicalreasoning - legalops - intent - functioncalling - structured The base Qwen…
An archived, matched pair of RG-Diff and SC-AR adapters for FlexDLLM: Diffuse When Possible, Autoregress When Necessary. The two stages run on a shared LLaDA-8B-Instruct backbone. RG-Diff emits during diffusion; SC-AR generates a short autoregressive repair, and diffusion resumes. This is a research preview of the existing math-oriented checkpoint pair. The release has passed a real-model ASK/repair/resume smoke test. The paper's full benchmark scores have not been reproduced for this bundle. The download contains approximately 521.5 MiB of adapters, tokenizer and configuration. The 8B backbone is downloaded separately by the loader. Both adapters use LoRA rank 32, alpha 64. RG-Diff is…
Questions
Which Text generation models are most downloaded?
By monthly downloads reported by the Hugging Face Hub: .