mlsanityData to 13 Oct 2026

T-pro 2.0: An Efficient Russian Hybrid-Reasoning Model and Playground

arXiv 2512.10430 ↗First posted 11 Dec 2025cs.CLcited by 2

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

We introduce T-pro 2.0, an open-weight Russian LLM for hybrid reasoning and efficient inference. The model supports direct answering and reasoning-trace generation, using a Cyrillic-dense tokenizer and an adapted EAGLE speculative-decoding pipeline to reduce latency. To enable reproducible and extensible research, we release the model weights, the T-Wix 500k instruction corpus, the T-Math reasoning benchmark, and the EAGLE weights on Hugging Face. These resources allow users to study Russian-language reasoning and to extend or adapt both the model and the inference pipeline. A public web demo exposes reasoning and non-reasoning modes and illustrates the speedups achieved by our inference stack across domains. T-pro 2.0 thus serves as an accessible open system for building and evaluating efficient, practical Russian LLM applications.

Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 46 of its 72 references that are in the map. First posted on arXiv 11 Dec 2025.

Builds on

46 of 72 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Qwen3 Technical Report14 May
  2. 2024DeepSeek-V3 Technical Report27 Dec

Other references · 44

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,093
  2. 2017Proximal Policy Optimization Algorithmscited by 14,592
  3. 2024The Llama 3 Herd of Modelscited by 14,486
  4. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,986
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,247
  7. 2025Qwen2.5-VL Technical Reportcited by 5,499
  8. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,270
  9. 2024Qwen2.5 Technical Reportcited by 4,840
  10. 2024GPT-4o System Cardcited by 4,536
  11. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,693
  12. 2024Qwen2 Technical Reportcited by 2,102
32 more
  1. 2025Gemma 3 Technical Reportcited by 1,748
  2. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,625
  3. 2023Accelerating Large Language Model Decoding with Speculative Samplingcited by 990
  4. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertaintycited by 569
  5. 2024From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipelinecited by 491
  6. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Modelscited by 461
  7. 2024Better & Faster Large Language Models via Multi-token Predictioncited by 350
  8. 2023Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicinecited by 324
  9. 2024Skywork-Reward: Bag of Tricks for Reward Modeling in LLMscited by 324
  10. 2025EAGLE-3: Scaling up Inference Acceleration of Large Language Models via Training-Time Testcited by 319
  11. 2023From Quantity to Quality: Boosting LLM Performance with Self-Guided Data Selection for Instruction Tuningcited by 303
  12. 2023SemDeDup: Data-efficient learning at web-scale through semantic deduplicationcited by 300
  13. 2023Language Model Tokenizers Introduce Unfairness Between Languagescited by 212
  14. 2025Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergycited by 207
  15. 2025Llama-Nemotron: Efficient Reasoning Modelscited by 155
  16. 2025A Minimalist Approach to LLM Reasoning: from Rejection Sampling to Reinforcecited by 143
  17. 2022mGPT: Few-Shot Learners Go Multilingualcited by 141
  18. 2025RewardBench 2: Advancing Reward Model Evaluationcited by 141
  19. 2024Non-Determinism of "Deterministic" LLM Settingscited by 92
  20. 2025AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learningcited by 85
  21. 2025HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languagescited by 74
  22. 2025Part I: Tricks or Traps? A Deep Dive into RL for LLM Reasoningcited by 54
  23. 2023A Family of Pretrained Transformer Language Models for Russiancited by 27
  24. 2024MERA: A Comprehensive LLM Evaluation in Russiancited by 22
  25. 2025PairJudge RM: Perform Best-of-N Sampling with Knockout Tournamentcited by 22
  26. 2025On the Effect of Sampling Diversity in Scaling LLM Inferencecited by 18
  27. 2024Vikhr: The Family of Open-Source Instruction-Tuned Large Language Models for Russiancited by 12
  28. 2024Facilitating large language model Russian adaptation with Learned Embedding Propagationcited by 8
  29. 2024Can DPO Learn Diverse Human Values? A Theoretical Scaling Lawcited by 7
  30. 2025Investigating the Zone of Proximal Development of Language Models for In-Context Learningcited by 6
  31. 2019An Analisys of Application Logs with Splunk : developing an App for the synthetic analysis of data and security incidentscited by 1
  32. 2025DOoM: Difficult Olympiads of Mathcited by 1

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mixcited by 1
  2. 2026Learning Selective LLM Autonomy from Copilot Feedback in Enterprise Customer Support Workflowscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning
  2. 2026Tandem: Riding Together with Large and Small Language Models for Efficient Reasoning
  3. 2025Reasoning Beyond Limits: Advances and Open Problems for LLMs
  4. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
  5. 2025Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging -- An Open Recipe
  6. 2026Domyn-Small: A European 10B Reasoning Language Model
  7. 2024Enhancing Reasoning Capabilities of LLMs via Principled Synthetic Logic Corpus
  8. 2024Inference Scaling vs Reasoning: An Empirical Analysis of Compute-Optimal LLM Problem-Solving