Catalog Systems Memory benchmarks & evaluation OSRS Bench (Old School RuneScape agent benchmark)

OSRS Bench (Old School RuneScape agent benchmark)

https://github.com/grahamannett/osrs-bench

Community benchmark suite for LLM agents playing Old School RuneScape, a sandbox MMORPG with deep partial-observability, skill-grinding loops, and economy interaction. Tests long-horizon planning (hour-to-day quests), tool inventory management, and visual UI comprehension. Community-built; not affiliated with Jagex.

Inactive. development has stopped

Inactive · Verified 2026-05-14 · fresh as of 2026-05-14 · 25 cited facts, 0 with a third-party source · 1 recorded change · Suggest a correction

At a glance

Type
Long-horizon MMO agent benchmark — Old School RuneScape repo fix
Tier
T4
Section
Memory benchmarks & evaluation
Created
2024 (community project) repo fix
Latest release
not applicable — not OSS fix
License
MIT (assumed — community repo, no explicit LICENSE file checked) repo fix
GitHub
small community repo (<50★), Python repo fix
Pricing
not applicable — open benchmark; free to use repo fix
Funding
not applicable — not commercial fix
Freshness
2026-05-14 fix
Layer
agent-infra repo fix
Licensing
open-source repo fix

Memory taxonomy

storage
n/a
retrieval
n/a
persistence
n/a
update
read-only
unit
n/a
governance
n/a
conflict
n/a

When to use

Optimised for: not applicable - eval dataset, not a system

Anti-fit: not applicable - eval dataset, not a system

Pros & cons

Pros

One of the few long-horizon (multi-hour to multi-day) agentic benchmarks built around a real, deeply-engineered game world; high partial-observability and economic loops.

Cons

Niche community benchmark with no standardized leaderboard; reproducibility hampered by Jagex anti-bot policies (RuneScape EULA forbids automation); not peer-reviewed.

Claims & capabilities

Used by community as exemplar long-horizon agent task — quest completion times measured in hours; full account progression measured in months of agent runtime

Technical surface

API surface
not applicable — eval dataset, not a system
Backend storage
not applicable — eval dataset, not a system
Deployment
not applicable — run locally
Embedding model
not applicable — eval dataset, not a system
Multi-tenancy
not applicable — eval dataset, not a system
MCP
not applicable — benchmark / evaluation harness
A2A
not applicable — benchmark / evaluation harness
OpenTelemetry
not applicable — benchmark / evaluation harness

Similar systems

Other memory benchmarks & evaluation in the catalog, ranked by inbound references.

  • LoCoMo T3

    Conversations spanning up to 35 sessions / 300 turns / ~9K tokens, built from LLM-agent personas grounded on temporal event graphs and verified by human annotators. Tests QA (single-hop, multi-hop, temporal, commonsense, adversarial), event-graph summarisation, multimodal dialogue.

  • GAIA T3

    Benchmark for general AI assistants made of real-world questions that require reasoning, multimodality, web browsing and tool use, with a public leaderboard on Hugging Face. Traction: ~917 likes on the Hugging Face dataset

  • LongBench T3

    Tsinghua THUDM. 21 datasets across 6 categories (single-doc QA, multi-doc QA, summarisation, few-shot ICL, synthetic, code completion) in English + Chinese. ~6.7K words English / 13.4K chars Chinese average. v2 at ACL 2025 with harder, longer instances.

  • LongMemEval T3

    5 core memory abilities — info extraction, multi-session reasoning, temporal reasoning, knowledge updates, abstention — via 500 curated questions in synthetic chat histories. Two scales: LongMemEval-S (115K tokens), LongMemEval-M (up to 1.5M).

  • RULER T3

    NVIDIA. Extends NIAH with multi-needle retrieval, multi-hop entity tracing, aggregation, QA — configurable for arbitrary length / difficulty. Evaluated 17 long-context models on 13 tasks.

  • InfiniteBench (∞Bench) T3

    Tsinghua / OpenBMB. First benchmark with average input >100K tokens. 12 tasks across English + Chinese: long novel QA, code debugging, math in long context, summarisation, retrieval. Realistic documents (novels, code) rather than synthetic filler.

Sources (24 cited facts, 0 third-party)

Each fact links to the page it came from; the badge says who published it (how to read sources).

What changed

Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)

  1. 2026-10-22
    • Lifecycle first recorded as “unknown”
    #195