RRepoGEO

REPOGEO REPORT · LITE

tinkoff-ai/CORL

Default branch main · commit 6afec904 · scanned 6/21/2026, 5:32:26 PM

GitHub: 1,362 stars · 171 forks

Scan history for this repo

Score trend below includes all ready runs (older left, newer right; scroll horizontally if needed). The table is collapsed by default—expand for newest-first rows, 10 per page.

Score trend (left → right: older → newer)

2 ready scans. Expand the table below for newest-first rows (10 per page, paginated).

AI VISIBILITY SCORE
40 /100
Critical
Category recall
0 / 2
Not recommended in any query
Rule findings
2 pass · 0 warn · 0 fail
Objective metadata checks
AI knows your name
3 / 3
Direct prompts that named your repo
HOW TO READ THIS REPORT

Action plan is what to do next — copy-pasteable changes prioritized by impact. Category visibility is the real GEO test: when a user asks an AI a brand-free question that should surface tinkoff-ai/CORL, does the AI actually recommend you — or your competitors? Objective checks verify the metadata signals AI engines weight first. Self-mention check detects whether AI even knows you exist by name.

Action plan — copy-paste fixes

3 prioritized changes generated by gemini-2.5-flash. Mark items done after you ship the fix.

OVERALL DIRECTION
  • highreadme#1
    Reposition the core value proposition in the README's opening

    Why:

    CURRENT
    # CORL (Clean Offline Reinforcement Learning)

[](https://twitter.com/vladkurenkov/status/1669361090550177793)
[](https://arxiv.org/abs/2210.07105)
[](https://github.com/tinkoff-ai/CORL/blob/main/LICENSE)
[](https://github.com/astral-sh/ruff)

🧵 CORL is an Offline Reinforcement Learning library that provides high-quality and easy-to-follow single-file implementations of SOTA ORL algorithms.
    COPY-PASTE FIX
    # CORL (Clean Offline Reinforcement Learning): A Benchmarking Framework for SOTA ORL Algorithms

CORL is a research-friendly Offline Reinforcement Learning library providing high-quality, easy-to-follow single-file implementations of State-of-the-Art (SOTA) ORL algorithms, designed for robust benchmarking and experimentation.
  • mediumreadme#2
    Add a 'Comparison to other libraries' section in the README

    Why:

    COPY-PASTE FIX
    ## Comparison to other libraries

While inspired by [CleanRL](https://github.com/vwxyzjn/cleanrl) for its single-file, high-quality online RL implementations, CORL specifically focuses on **Offline Reinforcement Learning**. Unlike broader frameworks like [RLlib](https://github.com/ray-project/ray) or [DeepMind's Acme](https://github.com/deepmind/acme) which offer extensive features for various RL paradigms, CORL prioritizes providing highly reproducible, single-file implementations of SOTA offline RL algorithms for research and benchmarking. [Stable Baselines3](https://github.com/DLR-RM/stable-baselines3) is a popular choice for online RL, whereas CORL fills the gap for robust offline policy learning.
  • lowabout#3
    Refine the repository description to emphasize benchmarking and research

    Why:

    CURRENT
    High-quality single-file implementations of SOTA Offline and Offline-to-Online RL algorithms: AWAC, BC, CQL, DT, EDAC, IQL, SAC-N, TD3+BC, LB-SAC, SPOT, Cal-QL, ReBRAC
    COPY-PASTE FIX
    A research-friendly library providing high-quality, single-file implementations of State-of-the-Art (SOTA) Offline and Offline-to-Online RL algorithms, designed for robust benchmarking and experimentation.

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

Category visibility — the real GEO test

Brand-free queries asked to google/gemini-2.5-flash. Did AI recommend you, or someone else?

Same questions for every model — switch tabs to compare answers and rankings.

Recall
0 / 2
0% of queries surface tinkoff-ai/CORL
Avg rank
—
Lower is better. #1 = top recommendation.
Share of voice
0%
Of all named tools, what % are you?
Top rival
ray-project/ray
Recommended in 2 of 2 queries
COMPETITOR LEADERBOARD
  1. ray-project/ray · recommended 2×
  2. pytorch/pytorch · recommended 2×
  3. vwxyzjn/cleanrl · recommended 2×
  4. deepmind/acme · recommended 1×
  5. DLR-RM/stable-baselines3 · recommended 1×
  • CATEGORY QUERY
    What are the best libraries for implementing state-of-the-art offline reinforcement learning algorithms?
    you: not recommended
    AI recommended (in order):
    1. Acme (deepmind/acme)
    2. RLlib (ray-project/ray)
    3. Stable Baselines3 (DLR-RM/stable-baselines3)
    4. D4RL (rail-berkeley/d4rl)
    5. PyTorch (pytorch/pytorch)
    6. TensorFlow (tensorflow/tensorflow)
    7. CleanRL (vwxyzjn/cleanrl)

    AI recommended 7 alternatives but never named tinkoff-ai/CORL. This is the gap to close.

    Show full AI answer
  • CATEGORY QUERY
    Seeking a clean, single-file framework for benchmarking offline reinforcement learning algorithms and experiments.
    you: not recommended
    AI recommended (in order):
    1. MiniHack (MiniHack-RL/minihack)
    2. CleanRL (vwxyzjn/cleanrl)
    3. RLlib (ray-project/ray)
    4. NumPy (numpy/numpy)
    5. JAX (google/jax)
    6. PyTorch (pytorch/pytorch)
    7. Gymnasium (Farama-Foundation/Gymnasium)

    AI recommended 7 alternatives but never named tinkoff-ai/CORL. This is the gap to close.

    Show full AI answer

Objective checks

Rule-based audits of metadata signals AI engines weight most.

  • Metadata completeness
    pass

  • README presence
    pass

Self-mention check

Does AI even know your repo exists when asked about it directly?

  • Compared to common alternatives in this category, what is the core differentiator of tinkoff-ai/CORL?
    pass
    AI named tinkoff-ai/CORL explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • If a team adopts tinkoff-ai/CORL in production, what risks or prerequisites should they evaluate first?
    pass
    AI named tinkoff-ai/CORL explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • In one sentence, what problem does the repo tinkoff-ai/CORL solve, and who is the primary audience?
    pass
    AI named tinkoff-ai/CORL explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

Embed your GEO score

Drop this badge into the README of tinkoff-ai/CORL. It auto-updates whenever the report is rescanned and links back to the latest report — easy public proof that you care about AI discoverability.

RepoGEO badge previewLive preview
MARKDOWN (README)
[![RepoGEO](https://repogeo.com/badge/tinkoff-ai/CORL.svg)](https://repogeo.com/en/r/tinkoff-ai/CORL)
HTML
<a href="https://repogeo.com/en/r/tinkoff-ai/CORL"><img src="https://repogeo.com/badge/tinkoff-ai/CORL.svg" alt="RepoGEO" /></a>
Pro

Subscribe to Pro for deep diagnoses

tinkoff-ai/CORL — Lite scans stay free; this card itemizes Pro deep limits vs Lite.

  • Deep reports10 / month
  • Brand-free category queries5 vs 2 in Lite
  • Prioritized action items8 vs 3 in Lite