AI

Language models trained from scratch, more than two dozen others fine-tuned, and tools for working with AI agents.

Argonne base models
8
models on Hugging Face
48
datasets
47
model downloads
18,912
dataset downloads
45,029

Argonne: language models trained from scratch

Each point is a base model, placed by the date its Hugging Face repository was created and by its size; the larger the point, the more its family of checkpoints has been downloaded.

250M 500M 1B 2B 5B Jul 2025 Jan 2026 Jul 2026 Argonne 1.0: 276M parameters, on Hugging Face since March 3, 2025, 317 downloads across 2 checkpoints 1.0 276M Argonne 1.5: 357M parameters, on Hugging Face since March 14, 2025, 187 downloads across 1 checkpoints 1.5 357M Argonne 2.0: 4.92B parameters, on Hugging Face since January 24, 2026, 150 downloads across 1 checkpoints 2.0 4.92B Argonne 2.5: 1.27B parameters, on Hugging Face since April 5, 2026, 3,042 downloads across 5 checkpoints 2.5 1.27B Argonne 3.0: 2.88B parameters, on Hugging Face since June 1, 2026, 2,820 downloads across 3 checkpoints 3.0 2.88B Argonne 3.5: 2.88B parameters, on Hugging Face since August 1, 2026, 533 downloads across 2 checkpoints 3.5 2.88B Argonne 4.0: 1.04B parameters, on Hugging Face since August 13, 2026, 784 downloads across 2 checkpoints 4.0 1.04B Argonne 4.5: 2.06B parameters, on Hugging Face since September 29, 2026, 2,127 downloads across 5 checkpoints 4.5 2.06B

Fine-tuned models

27 models built on other families' checkpoints, by what the fine-tuning taught them; each bar follows a model's downloads to date, on one scale.

  • Reasoning 10
  • Task models, 2024 8
  • Preference optimization 5
  • Tool calling 2
  • Format conversion 1
  • Chat 1

Reasoning

10 models

small models taught to think before they answer

  1. Argonne-Qwen1.5-0.5B-think from Qwen1.5-0.5B, Jul 2026 754
  2. Qwen3.5-0.8B-Opus-4.6-thinking from Qwen3.5-0.8B-thinking, Mar 2026 95
  3. Qwen3.5-0.8B-thinking from Qwen3.5-0.8B-Base, Mar 2026 204
  4. llama3.2-3b-thinking from Llama-3.2-3B, Mar 2026 144
  5. Qwen3-0.6b-thinking from Qwen3-0.6B, Feb 2026 1,803
  6. llama-3-2-1b-open-r1-mot-sft from Llama-3.2-1B, Nov 2025 165
  7. llama3.2-1b-thinking from Llama-3.2-1B, Nov 2025 2,432
  8. qwen2.5-0.5b-open-r1-mot-cot-sft from Qwen2.5-0.5B, Nov 2025 82
  9. finetuned-llama-3.2-3b-math-reasoning from Llama-3.2-3B, Nov 2025 47
  10. Qwen2.5-1.5B-Instruct-Lora-Deepseek-R1 from Qwen2.5-1.5B-Instruct, Mar 2025 115

Task models, 2024

8 models

question answering, sentiment, summaries and named entities

  1. roberta-large-ner from roberta-large, Jul 2024 398
  2. bert-base-ner from bert-base-cased, Jul 2024 59
  3. distilbert-base-cased-ner from distilbert-base-cased, Jul 2024 59
  4. t5-base-summary-model from t5-base, Jul 2024 51
  5. t5-large-summary-model from t5-large, Jul 2024 60
  6. finetuned_qa_squad_bert_base_model from bert-base-uncased, Jul 2024 277
  7. finetuned_sentiment_bert_base_model from bert-base-uncased, Jul 2024 1,177
  8. bert-base-qa-finetuned-squad-v2 from bert-base-uncased, Jun 2024 50

Curated and synthetic data

47 datasets on Hugging Face: cleaned pretraining text, answers and reasoning traces written by large models for training small ones, summaries, captions and preference pairs. Each bar follows a dataset's downloads to date, on one scale.

Which model wrote which

Most of the instruction sets come from one design: the same public prompts, answered by each of five large models. Darker cells were downloaded more.

OpenAssistantSynthetic promptsOpenR1 MathAlpacaDolly 15kUltraFeedback
Llama 4 Maverick 219 234 244 225 225
Llama 4 Scout 206 202 223 227 199
Llama 3.3 70B 169 183 210 220 252
Llama 3.3 8B 156 185 219 206 215
Mistral Large 227 156 244

Reasoning traces

13 datasets

worked solutions and chains of thought, for teaching models to think

  1. MiniMax-M2.1-Mixture-of-Thoughts 100K–1M rows, Dec 2025 6,276
  2. 0.5M-thinking 100K–1M rows, Jan 2026 5,040
  3. openmath-reasoning-medley 1M–10M rows, Feb 2026 4,781
  4. 0.9M-thinking 100K–1M rows, Jan 2026 4,350
  5. gsm8k-thinking 1K–10K rows, Dec 2025 1,564
  6. arxiv-qa-thinking 100K–1M rows, Jan 2026 1,177
  7. toucan-agentic-thinking 100K–1M rows, Jan 2026 1,044
  8. gsm8k-Llama-4-Maverick-17B-128E-Instruct-FP8 1K–10K rows, Jun 2025 500
All 13
  1. s1k-magistral-small-2506 1K–10K rows, Jun 2025 246
  2. Llama-4-Maverick-17B-128E-Instruct-FP8-OpenR1-Math-220k 10K–100K rows, May 2025 244
  3. Llama-4-Scout-17B-16E-Instruct-FP8-OpenR1-Math-220k 10K–100K rows, May 2025 223
  4. Llama-3.3-8B-Instruct-OpenR1-Math-220k 10K–100K rows, May 2025 219
  5. Llama-3.3-70B-Instruct-OpenR1-Math-220k 10K–100K rows, May 2025 210

Instruction responses

21 datasets

answers written by large models to public prompt sets, for fine-tuning small ones

  1. ProcessedOpenAssistant 10K–100K rows, Feb 2025 429
  2. mistral-awesome-chatgpt-prompts under 1K rows, Jun 2025 285
  3. Llama-3.3-70B-Instruct-databricks-dolly-15k 10K–100K rows, May 2025 252
  4. Llama-4-Maverick-17B-128E-Instruct-FP8-instruct-synthetic-prompt-responses 10K–100K rows, May 2025 234
  5. ProcessedOpenAssistant-mistral-large-2411 10K–100K rows, Apr 2025 227
  6. Llama-4-Scout-17B-16E-Instruct-FP8-alpaca-cleaned 10K–100K rows, May 2025 227
  7. Llama-4-Maverick-17B-128E-Instruct-FP8-alpaca-cleaned 10K–100K rows, May 2025 225
  8. Llama-4-Maverick-17B-128E-Instruct-FP8-databricks-dolly-15k 10K–100K rows, May 2025 225
All 21
  1. Llama-3.3-70B-Instruct-alpaca-cleaned 10K–100K rows, May 2025 220
  2. Llama-4-Maverick-17B-128E-Instruct-FP8-ProcessedOpenAssistant 10K–100K rows, May 2025 219
  3. Llama-3.3-8B-Instruct-databricks-dolly-15k 10K–100K rows, May 2025 215
  4. Llama-4-Scout-17B-16E-Instruct-FP8-Instruct-FP8-ProcessedOpenAssistant 10K–100K rows, May 2025 206
  5. Llama-3.3-8B-Instruct-alpaca-cleaned 10K–100K rows, May 2025 206
  6. Llama-4-Scout-17B-16E-Instruct-FP8-instruct-synthetic-prompt-responses 10K–100K rows, May 2025 202
  7. Llama-4-Scout-17B-16E-Instruct-FP8-databricks-dolly-15k 10K–100K rows, May 2025 199
  8. Llama-3.3-8B-Instruct-instruct-synthetic-prompt-responses 10K–100K rows, May 2025 185
  9. Llama-3.3-70B-Instruct-instruct-synthetic-prompt-responses 10K–100K rows, May 2025 183
  10. Llama-3.3-70B-Instruct-ProcessedOpenAssistant 10K–100K rows, May 2025 169
  11. llama-awesome-chatgpt-prompts under 1K rows, Jun 2025 158
  12. instruct-synthetic-prompt-responses-mistral-large-2411 10K–100K rows, Apr 2025 156
  13. Llama-3.3-8B-Instruct-ProcessedOpenAssistant 10K–100K rows, May 2025 156

Pretraining text

1 dataset

web text cleaned for training from scratch

  1. processed-fineweb-edu 1M–10M rows, Feb 2025 1,371

Code

1 dataset

seeds for generating programming data

  1. oss-code-seeds 100K–1M rows, Feb 2026 247

Data that trained the models here

As the models' own cards declare them.

Every model, on one calendar

One circle per model, by the date its repository was created; the area follows its downloads to date.

Agents and tools

Sage

Ask your docs a question. Get an answer that links the exact section it came from.

Started
December 2025

claudekeeper

Schedule prompts for Claude Code, and keep your sessions running through the 5-hour usage cap.

Started
July 2026
Version
0.3.1, 4 releases

RedLLM

A GPT-style model pretrained on the classic novel Dream of the Red Chamber (红楼梦).

Started
February 2025

hpcpilot, an agent for Slurm clusters, is on the HPC page.

Articles on language models and agents

  1. Evaluating AI Agents When Every Run Differs: Repeated Trials, Paired Comparisons and Graders
  2. Tool Calling in Language Model Agents: Designing the Interface Between a Model and Its Tools
  3. Workflow or Agent: Deciding Which Steps Belong in Code and Which Belong to the Model
  4. Why Temperature Zero Is Not Reproducible: Floating-Point Arithmetic and Batching in Language Model Inference
  5. Tuning the GPU Interconnect in Multi-Node Language Model Pretraining
  6. A Desktop Grace Blackwell Machine as Research Infrastructure: Serving and Training Benchmarks for the NVIDIA DGX Spark
  7. Evaluating an Agentic System: Three Axes, and What Auditing the Measurements Changed
  8. Argonne 3.5-base: Retraining an Unchanged Architecture With a Revised Recipe
  9. Argonne 3.5-think: From Base Model to Reasoning Model
  10. Pretraining a Language Model From Scratch: Argonne 1.0 to 3.0
  11. How I Taught a Small Language Model to Reason

Figures as of November 19, 2026: models, sizes and download counts from Hugging Face; Argonne sizes for 1.0 to 3.0 as the pretraining article reports them.