
GITNUXSOFTWARE ADVICE
Sports RecreationTop 10 Best Chess Training Software of 2026
Top 10 chess training software ranking with lessons, analysis, and tools, comparing DecodeChess, Lichess, and Chess.com for study needs.
How we ranked these tools
Core product claims cross-referenced against official documentation, changelogs, and independent technical reviews.
Analyzed video reviews and hundreds of written evaluations to capture real-world user experiences with each tool.
AI persona simulations modeled how different user types would experience each tool across common use cases and workflows.
Final rankings reviewed and approved by our editorial team with authority to override AI-generated scores based on domain expertise.
Score: Features 40% · Ease 30% · Value 30%
Gitnux may earn a commission through links on this page — this does not influence rankings. Editorial policy
DecodeChess is the best pick for coaches who want repeatable engine-evaluated lesson workflows tied to review and annotation, while Lichess is the cheapest entry for solo learners or small groups needing shareable studies and fast analysis, and if you prefer spaced study loops on your own, Chessable fits repertoire-focused training.
Editor’s top 3 picks
Three quick recommendations before you dive into the full comparison below — each one leads on a different dimension.
DecodeChess
Lesson authoring that links engine evaluation to task prompts and blunder-style review in one workflow.
Built for fits when coaches need repeatable engine-evaluated lesson workflows tied to review and annotation..
Lichess
Editor pickInteractive studies that convert imported games into step-by-step, click-through lessons with shareable structure.
Built for fits when solo learners or small coaching groups need shareable studies and fast engine review..
Chess.com
Editor pickBlunder review mode that filters post-game analysis to the moves that most hurt accuracy.
Built for fits when individual players want daily tactics plus engine-backed analysis without separate tools..
Related reading
Comparison Table
DecodeChess
vertical specialistChess engine that explains its evaluations in natural language.
Lesson authoring that links engine evaluation to task prompts and blunder-style review in one workflow.
DecodeChess is oriented around engine evaluation-driven training and lesson sequences rather than standalone puzzles. Lesson pages can attach analysis boards, move-by-move commentary, and task prompts that map to what the engine flags. The workflow supports PGN import for reusing existing games and FEN setup for creating controlled test positions. Review sessions focus on blunder review style feedback and move annotation panels so the training session stays grounded in concrete candidate lines.
A tradeoff is that advanced drill coverage depends on how lessons and tasks are authored, so teams seeking fully automatic coverage may need custom content. DecodeChess fits when a coach wants repeatable study plans for specific openings or recurring tactical themes, with review sessions tied to engine judgments. It also fits solo players who want to convert their own games into structured practice with consistent review prompts.
- +Engine-backed lesson authoring with review prompts tied to mistakes
- +PGN import and FEN setup support controlled, repeatable training sessions
- +Move annotation panel keeps coaching context visible during review
- +Engine line review helps connect candidate moves to player decisions
- –Advanced coverage requires manual lesson and task design
- –Deep customization can slow down setup for one-off study needs
- –Automated personalization is limited without curated lesson structure
- –Complex lesson trees can feel heavy for short sessions
Chess coaches
Create review-driven lessons from student games
Consistent coaching across sessions
Serious self-study players
Convert a lost game into drills
Targeted fix for recurring errors
Show 1 more scenario
Opening builders
Train repertoire-specific patterns and mistakes
Cleaner decision-making under pressure
Structure lessons around critical move choices and annotate candidate lines for review.
Best for: Fits when coaches need repeatable engine-evaluated lesson workflows tied to review and annotation.
More related reading
Lichess
vertical specialistFree open-source chess server with puzzles, studies, and engine analysis.
Interactive studies that convert imported games into step-by-step, click-through lessons with shareable structure.
Lichess provides an analysis board for board-first training that supports candidate lines, blunder review after a game, and engine-driven positional evaluation during review. Studies allow structured lesson content with move-linked diagrams, bookmarks, and collaboration through shareable study pages. Puzzle training includes multiple formats such as tactics, timed rush play, and puzzle sequences designed for pattern repetition.
A key tradeoff is that Lichess automation and administration are limited compared with dedicated chess academies that manage cohorts, rosters, and enforced practice plans. Lichess fits best when an individual coach or self-directed learner needs engine-backed review and shareable lesson material without building a separate LMS.
- +Engine-backed analysis with clear variation review and move-level annotation
- +Interactive studies that turn PGN lines into lesson-like sequences
- +Tactical puzzle modes with timed and streak-oriented practice
- +Large public game pool for opening and move-order benchmarking
- –Team-style governance and cohort automation are not a focus
- –Endgame training relies on general analysis tools instead of dedicated drills
- –Advanced repertoire building requires manual curation and study organization
- –Workflow customization is thinner than purpose-built training platforms
Solo learners
After-game engine review and improvement
Fewer repeated mistakes
Coaches
Create and share training study content
Consistent student practice
Show 2 more scenarios
Club trainers
Opening and move-order comparisons
Sharper opening choices
Use public games and analysis to compare variations and refine recommendations for students.
Tactical improvers
Timed pattern repetition practice
Higher tactical consistency
Train with puzzle rush formats to improve speed and accuracy on common tactical motifs.
Best for: Fits when solo learners or small coaching groups need shareable studies and fast engine review.
Chess.com
vertical specialistFull-featured chess platform with lessons, puzzles, game review, and live play.
Blunder review mode that filters post-game analysis to the moves that most hurt accuracy.
Chess.com is distinct in how it ties practice directly to gameplay, with puzzles, tactics drills, and annotated analysis connected to the same move interface. Study sessions can start from a custom position via FEN setup, then use the analysis board to review candidate moves with multi-line engine output. PGN import and game uploads support structured review for opening choices and tactical themes, with a blunder review view that concentrates attention on key errors.
A tradeoff appears in automation and programmatic control, since Chess.com provides no documented admin provisioning, RBAC, or public API surface for custom training pipelines. Chess.com fits best when individual users or small clubs want guided tactics and analysis with minimal tooling overhead, rather than when teams need dataset-level integration.
- +Puzzle Rush and daily puzzles create repeatable tactical practice loops
- +Blunder review highlights swing moves with focused post-game analysis
- +Analysis board supports FEN setup and PGN import for targeted study
- +Engine variations make candidate-move review practical for study sessions
- –No documented public API limits custom training automation and integrations
- –Engine analysis is account-centric, which reduces portability to local workflows
- –Multi-step study building can feel less structured than dedicated course platforms
- –Limited governance controls for larger organizations constrain club administration
Solo improvers
Tactics drilling tied to daily practice
Faster tactical decision-making
Club coaches
Annotate games for group study
Clearer shared review sessions
Show 2 more scenarios
Opening repertoire builders
Position-based testing of variations
More consistent opening choices
Set positions with FEN and test candidate continuations using multi-line engine analysis.
Tournament prep players
Targeted review before key games
Reduced repeated mistake rate
Analyze recent games to identify recurring blunders and focus study on the causing positions.
Best for: Fits when individual players want daily tactics plus engine-backed analysis without separate tools.
More related reading
ChessBase
enterpriseProfessional chess database and analysis software for serious training.
Endgame tablebase probe inside the study workflow to verify forced endgame sequences beyond engine evaluation.
ChessBase is a long-running desktop chess training suite built around deep analysis, study workflows, and database-driven preparation. It supports engine integration via the UCI protocol, PGN import and position setup for analysis boards, and multi-variation review with clear move annotation panels.
Training value comes from opening tree explorer and endgame tablebase probe when paired with compatible engines and saved study positions. For structured practice, its puzzle and repertoire workflows fit players who review games repeatedly in a consistent local environment.
- +Tight engine integration through UCI so analysis stays consistent across sessions
- +Strong PGN import and analysis board workflows for post-game review
- +Endgame tablebase probe support improves endgame verification beyond engine search
- +Database-first navigation helps repeat study on openings and lines
- –Learning curve is higher than guided puzzle-first trainers
- –Setup and tuning of engines affects analysis throughput and result quality
- –Automation surface is limited compared with tools built around scripted trainers
- –Spaced repetition style training is less central than manual review and study
Best for: Fits when serious players want local, database-driven analysis and repeatable study positions.
Shredder Chess
vertical specialistCommercial chess engine with built-in adaptive training puzzles.
Blunder review mode that routes engine findings into targeted replays focused on repeat mistakes.
Shredder Chess focuses on chess training workflows that turn analysis and preparation into repeatable drills. The tool supports game loading and position setup for engine-guided review, then routes results into structured study tasks like tactical practice and opening work.
Shredder Chess includes training modes that emphasize mistake-driven iteration and position-based repetition. The workflow centers on engine evaluation output and interactive boards for reviewing lines and candidate moves.
- +Engine-assisted review turns post-game analysis into targeted practice
- +Interactive move navigation supports candidate-line walkthroughs
- +Position setup and game import feed recurring training sessions
- +Tactical and opening practice can be structured around training goals
- –Setup of training focus areas can feel manual for some users
- –Advanced customization of study generation needs careful configuration
- –Dashboard-style progress analytics are limited compared with dedicated learning platforms
- –Some workflow steps depend on consistent PGN and position formatting
Best for: Fits when players want engine-guided review converted into repeatable drills.
Chess King
vertical specialistSuite of chess training apps covering tactics, endings, and openings.
Opening repertoire builder that ties line selection to repeated practice and post-game testing inside the study flow.
Chess King pairs an engine-driven training flow with an opening and endgame focus that supports structured study. The site emphasizes tactical practice, guided analysis after games, and repertoire building from annotated content.
It also supports position setup workflows via common chess formats so training material can start from specific FEN or importable game states. Engine analysis features target move-accuracy feedback and detailed review so errors like blunders are easy to trace into next drills.
- +Tactical drilling workflow that converts mistakes into follow-up practice
- +Clear game review with engine evaluation and annotation-driven feedback
- +Opening-focused training centered on building and testing repertoire lines
- +Uses import and position setup workflows for starting drills from specific states
- –Advanced study paths need consistent configuration to keep drills aligned
- –Collaboration and admin governance controls are limited for team use
- –API-based automation is not a primary integration surface for external systems
- –Multi-variation analysis depth is constrained by the in-site analysis workflow
Best for: Fits when a solo player wants engine-backed review, targeted tactics, and opening preparation in one study loop.
More related reading
Chess Magnet
vertical specialistWeb-based chess training system with adaptive lessons and puzzles.
Magnet-pattern tactics training that groups missed tactical cues into follow-up drill sets.
Chess Magnet is a chess training software built around magnet-style tactics and targeted practice drills rather than broad content libraries.
It focuses on creating repeatable training sessions from position sets and then tracking performance across practice runs.
The workflow centers on engine-aided review of moves and errors inside its training loop.
For users who want structured daily reps with clear feedback, it provides a tighter tactics-to-review cycle than many general-purpose study sites.
- +Magnet-based tactics drills turn missed patterns into focused repetition
- +Session-based training flow keeps engine review attached to practice mistakes
- +Compact interface supports fast start without building a complex syllabus
- +Performance tracking highlights which positions and move types recur
- –Fewer long-form repertoire tools than full opening and endgame study suites
- –Export and data portability options are limited for external analysis workflows
- –Advanced analysis controls lag behind apps with multi-PV and deep tuning
- –Custom training set creation requires manual curation of position inputs
Best for: Fits when tactics-focused study needs fast reps, mistake review, and minimal syllabus setup.
Chessable
vertical specialistSpaced-repetition chess course platform using MoveTrainer technology.
Spaced repetition lesson mode schedules recalls from exact annotated positions inside each course.
Chessable pairs structured lessons with a spaced repetition trainer built around move-by-move recall. The system uses annotated positions, deliberate practice prompts, and game-like boards to turn opening, middlegame, and endgame study into repeatable drills.
A repertoire builder workflow helps convert study into a usable plan, while post-session review highlights missed moves and concept gaps. Format and feedback are tightly coupled, so training sequences drive both the learning order and the evaluation loop.
- +Spaced repetition scheduling is embedded into lesson playback and drill flow
- +Move recall drills give immediate feedback with a board-first study loop
- +Repertoire builder turns course content into a concrete opening plan
- +Annotated content supports practical conversion from training positions to games
- –Deep customization is limited compared with DIY workflows in local training tools
- –Advanced analysis workflows depend on the platform’s study format rather than a general editor
- –PGN workflows are primarily study-ingestion focused rather than full post-game automation
- –Engine-driven investigation is constrained by lesson-oriented interfaces
Best for: Fits when repertoire-focused learners want guided spaced repetition and board-based drill feedback.
More related reading
Chessify
vertical specialistChess analysis platform offering cloud engines, opening databases, tablebases, and mobile study tools.
Blunder review mode ties each failed attempt to a corrective move path during post-attempt analysis.
Chessify turns study positions into a repeatable training workflow with interactive puzzles and guided review. The core loop centers on engine-assisted evaluation and move feedback so training sessions end with concrete correction targets.
Position setup and PGN-based import support lets users bring games and variations into the trainer for focused practice. Progress tracking ties session results back to recurring weaknesses for ongoing drill selection.
- +Engine-assisted feedback highlights specific inaccuracies after each attempt
- +PGN import supports moving from real games into targeted drills
- +Interactive puzzle sessions keep practice focused on concrete moves
- +Progress tracking makes it easier to spot repeat error patterns
- –Automation depth for large libraries is limited for team workflows
- –Advanced customization of puzzle selection is narrower than dedicated drill suites
- –Local engine control and multi-engine benchmarking are not designed for heavy users
- –Browser-based analysis can feel slow on very large PGN imports
Best for: Fits when individual players want engine-guided puzzle practice tied to game-based errors.
Lucas Chess
free/open-sourceFree desktop chess training program with configurable engines, tactical exercises, and skill-based practice.
Blunder review mode that pinpoints decisive errors with context from the analysis candidate tree.
Lucas Chess is a local-first chess training app built around a classic study workflow with engine-assisted analysis and annotated learning sessions. The software supports PGN import, FEN position setup, and board-based training that tracks accuracy by move and positions.
Its analysis boards provide candidate move trees and engine evaluation details that feed post-game review and blunder review mode. Rep from preparation to tactical drilling, Lucas Chess focuses on iterative practice rather than online coaching features.
- +PGN import and FEN setup support repeatable training from game databases
- +Engine analysis includes multi-PV candidate move lines for structured review
- +Blunder review mode highlights critical inaccuracies during post-game analysis
- +Tactical and positional drills are organized into repeatable practice sessions
- –Automation relies on local workflows and file-based inputs rather than APIs
- –Setup and tuning of engine strength needs manual attention for consistent benchmarks
- –Spaced repetition control is less granular than dedicated SRS trainers
- –Puzzle rush style pacing is limited compared with modern timed puzzle modes
Best for: Fits when local chess study workflows need engine-assisted analysis and repeatable practice loops.
Conclusion
After evaluating 10 sports recreation, DecodeChess stands out as our overall top pick — it scored highest across our combined criteria of features, ease of use, and value, which is why it sits at #1 in the rankings above.
Use the comparison table and detailed reviews above to validate the fit against your own requirements before committing to a tool.
How to Choose the Right chess training software
Chess training software in this guide covers engine-backed study loops for tactics, analysis, and repertoire practice, with workflow depth that ranges from interactive studies to local, database-driven analysis.
The lineup includes DecodeChess for engine-evaluated lesson authoring and blunder-style review, plus Lichess and Chess.com for study-style or blunder-focused training, and ChessBase and Lucas Chess for locally oriented PGN import and analysis workflows.
Chess training software for engine-evaluated lessons, study boards, and review workflows
Chess training software turns engine evaluation into training artifacts like guided lessons, interactive move trees, and post-game review sequences that connect mistakes to follow-up practice.
DecodeChess focuses on lesson authoring that links engine evaluation to task prompts and blunder-style review inside one workflow, while Chess.com centers on blunder review mode that filters post-game analysis to the moves that most hurt accuracy.
Lichess complements this with interactive studies that convert imported games into step-by-step click-through lessons, so the same PGN lines can become structured training sequences.
Across the tools covered here, the deciding differences are how evaluation is routed into drills, how imported games and positions are turned into repeatable study sessions, and how much control is available for automation and integration.
Engine-to-drill workflow, study conversion, and integration control
Chess training software only improves results when engine evaluation turns into training artifacts like lesson tasks, candidate-line walkthroughs, and mistake-tied replays. DecodeChess links engine evaluation to task prompts and blunder-style review in one workflow, so the same analysis step drives the next drill rather than stopping at a report.
The other deciding difference is how imported games and positions become repeatable training sessions. Lichess converts imported PGN lines into step-by-step interactive studies, while ChessBase routes PGN import into a study workflow with an endgame tablebase probe that stays inside the study flow.
Engine-evaluated lesson authoring tied to mistake review
DecodeChess connects engine evaluation to task prompts and blunder-style review so lesson creation and mistake practice share the same loop. Chessify and Shredder Chess also push blunder review into practice, but DecodeChess keeps that link inside lesson authoring rather than after-the-fact drill selection.
Interactive study conversion from imported games
Lichess turns imported games into click-through, step-by-step lessons using interactive studies, which makes PGN lines usable as training sequences. Chess.com also offers analysis and training loops, but it does not convert PGN studies into a shareable click-through structure as the primary workflow.
Blunder filtering focused on accuracy swing moves
Chess.com blunder review mode filters post-game analysis to the moves that hurt accuracy the most, which reduces review time spent on quiet inaccuracies. Shredder Chess similarly routes engine findings into targeted replays, but it organizes replay re-entry around specific repeat mistakes.
Local study workflow with endgame tablebase verification
ChessBase includes an endgame tablebase probe inside the study workflow, so forced endgame sequences can be verified beyond engine evaluation. DecodeChess focuses on engine-backed lesson workflows, while ChessBase emphasizes database-driven local analysis with tablebase validation.
Repertoire building that maps line selection to practice and testing
Chess King provides an opening repertoire builder that ties line selection to repeated practice and post-game testing inside the study flow. DecodeChess emphasizes engine-backed lesson creation tied to mistakes, while Chess King builds the study sequence around repertoire alignment.
Spaced repetition scheduling from annotated positions
Chessable embeds spaced repetition scheduling into lesson playback so recall is driven by exact annotated positions. Lucas Chess and ChessBase support PGN and engine-based analysis, but they do not schedule spaced recalls as a native lesson playback mechanism.
Choose by how analysis becomes practice and how much workflow control is available
Selection starts with the path engine evaluation takes before a user reaches a drill. DecodeChess routes evaluation into authored lesson tasks and blunder-style review, while Lichess routes imported PGN into interactive study steps that users click through.
The second axis is workflow control for automation and repeatability across sessions. Chess.com’s training loop stays account-centric without a documented public API, while Lucas Chess and ChessBase lean on local workflows and file-based inputs that reduce dependency on cloud orchestration.
Pick the route from engine output to the next training action
If engine evaluation must directly generate lesson tasks and mistake-tied prompts, DecodeChess fits because it links evaluation and blunder-style review inside lesson authoring. If imported games must become click-through training sequences, Lichess fits because interactive studies convert PGN lines into step-by-step lesson steps.
Decide whether training is replay-centered or study-conversion centered
For replay-centered training that returns to the moves that hurt accuracy most, Chess.com uses blunder review mode that filters the key swing moves. For study-conversion centered training, Lichess uses interactive studies so the lesson structure is created from the imported move lines rather than from post-game filters.
Validate endgame correctness inside the study workflow
If forced endgame sequences must be verified beyond engine evaluation, ChessBase includes an endgame tablebase probe within its study workflow. If endgame drills are secondary and mistake-linked tactics and lesson loops are primary, DecodeChess or Chess.com can match the workflow priority.
Match deployment and integration expectations to local vs platform-centric workflows
If local study workflows and file-based PGN and FEN inputs are acceptable, Lucas Chess and ChessBase support repeatable training from game databases. If an automation-centric integration path is required, Chess.com lacks a documented public API and keeps engine analysis account-centric, so integration depth is limited.
Align repertoire creation with how the platform records and tests line choices
If the workflow must convert opening line selection into repeated practice and post-game testing, Chess King is designed for repertoire builder to study alignment. If recall training must follow spaced repetition from exact annotated positions, Chessable schedules recalls inside lesson playback.
Check export and portability before committing to external toolchains
If exports and external portability matter for building additional analysis pipelines, prefer tools with broader data output than Chess Magnet because Chess Magnet has limited export and data portability for external workflows. If the training loop stays inside the platform, Chess Magnet’s magnet-pattern tactic drills can keep setup minimal for fast reps.
Who chess training software fits best
Different tools convert analysis into drills in different ways, so fit depends on the workflow at the moment mistakes are identified. Users who build custom lesson content and want engine-backed prompts tied to blunder-style review benefit from DecodeChess.
Players who prefer fast conversion of games into structured steps benefit from Lichess, while players who want tactical practice loops and swing-move filtering inside post-game review benefit from Chess.com.
Coaches and analysts who author repeated lesson tasks from engine results
DecodeChess supports engine-evaluated lesson authoring that links task prompts to blunder-style review so coaches can standardize how analysis becomes practice.
Solo players and small groups who want shareable PGN-to-study lesson structures
Lichess turns imported games into interactive studies that users can step through, which makes it practical for group review when the goal is consistent click-through lesson flow.
Players who prioritize daily tactics and accuracy swing detection after games
Chess.com uses Puzzle Rush and daily puzzles for repeatable tactics, and its blunder review mode filters the moves that most hurt accuracy so review stays focused.
Players who run local databases and need endgame validation inside the study workflow
ChessBase supports strong PGN import and analysis board workflows with an endgame tablebase probe in the study workflow for forced sequence verification.
Learners who want recall-driven training from exact annotated positions
Chessable schedules spaced repetition from annotated positions inside lesson playback, which targets move recall with immediate board-based feedback.
Common pitfalls when buying chess training software
Many buyers choose a tool based on analysis quality only, then discover the analysis does not convert into the training action format required. The second frequent failure is committing to the wrong workflow shape for imported games and repeatability.
Buyers also underestimate how engine integration decisions affect configuration and throughput, especially in local setups where engine tuning impacts results.
Choosing a platform that shows engine analysis but does not connect mistakes to the next drill in the same workflow
DecodeChess and Shredder Chess convert engine findings into targeted practice replays or prompts, while tools focused on general analysis can leave the user to manually build the follow-up drill loop.
Assuming interactive study conversion exists when the platform is primarily an account-centric training loop
Lichess converts PGN lines into interactive studies, while Chess.com keeps engine analysis account-centric and does not provide a documented public API for external training automation.
Overlooking setup and engine tuning work that affects analysis throughput
ChessBase requires engine setup and tuning that can change analysis throughput and result quality, and Lucas Chess needs manual engine strength tuning to keep benchmarks consistent across runs.
Buying for long-form repertoire building but only getting narrow drill coverage
Chess King is designed around repertoire building inside the study flow, while Chess Magnet emphasizes magnet-pattern tactics and has fewer long-form repertoire tools.
Selecting a spaced-repetition tool but expecting fully DIY customization like local editors
Chessable embeds spaced repetition scheduling inside its lesson playback format, while local training workflows like Lucas Chess and ChessBase allow more DIY structure but do not schedule recalls as a native lesson playback mechanism.
How We Selected and Ranked These Tools
We evaluated DecodeChess, Lichess, Chess.com, ChessBase, Shredder Chess, Chess King, Chess Magnet, Chessable, Chessify, and Lucas Chess by weighting features at 40%, then combining ease and value each at 30%. Features scoring favored engine-to-drill workflow quality like DecodeChess engine-backed lesson authoring that links evaluation to task prompts and blunder-style review, plus Lichess interactive study conversion from imported PGN.
Ease scoring favored setup paths that keep users moving from import and positions into review and practice. Value scoring favored repeatable training loops that reduce manual repetition work across sessions, and DecodeChess ranked highest because its lesson authoring pipeline and repeatable mistake-tied prompts reduce the setup time for coherent training sequences.
Frequently Asked Questions About chess training software
Which tools handle lesson authoring tied to engine evaluation and mistake prompts?
How does PGN and FEN import change the training setup workflow in chess training software?
When a learner needs multi-variation analysis with annotations, which platform provides the tightest review loop?
What breaks if a coach expects engine integration via UCI protocol across local and desktop environments?
Which tool is best when training must verify forced endgame lines beyond generic evaluation?
How do blunder review modes differ across chess training tools?
When a user needs spaced repetition scheduling tied to exact annotated positions, which software matches that requirement?
Which platforms support converting training from imported games into structured, shareable lessons?
What extensibility and automation options exist when admin controls and auditability are required?
Tools reviewed
Primary sources checked during evaluation.
Referenced in the comparison table and product reviews above.
Keep exploring
Comparing two specific tools?
Software Alternatives
See head-to-head software comparisons with feature breakdowns, pricing, and our recommendation for each use case.
Explore software alternatives→In this category
Sports Recreation alternatives
See side-by-side comparisons of sports recreation tools and pick the right one for your stack.
Compare sports recreation tools→