AI Agent Benchmarks — Saturation Catalog (agent edition · no-JS)

Machine-readable reference of agentic (agent = Model + Harness/scaffold) evaluation benchmarks and how saturated each is. Snapshot 2026-06. This page is built for AI agents and CLI browsers; it needs no JavaScript.

Machine endpoints: llms.txt (agent entry point) · data.json · data.csv · catalog.md · schema.json

saturation = SOTA as % of the metric ceiling (or human reference if no fixed ceiling): open(<50) · active(50-75) · near(75-90) · saturated(≥90). priority = 0.40·Headroom + 0.25·Durability + 0.20·Credibility + 0.15·Feasibility (1-5 each).

Caveat: Snapshot as of 2026-06. IMPORTANT: Anthropic's Claude Fable 5 (GA 2026-06-09) and Mythos 5 / Mythos Preview set or changed SOTA on several rows but were SUSPENDED 2026-06-12 under a US export-control directive — those entries are real-on-paper but the systems are not currently runnable (flagged in `system`/`notes`). Many SOTA figures are self-reported or aggregator-sourced; some cells are UNVERIFIED or PROVISIONAL and some public boards are stale. Treat as indicative, not definitive — verify against each row's `source`. 'agent' = Model + Harness/scaffold; the scaffold (named in `system`) materially affects scores.

Rows sorted by priority-to-beat. Full fields & per-benchmark notes are in data.json / catalog.md.

BenchmarkSaturationSOTA / who / whenHeadroomPrioritySource
Konwinski Prize (K Prize)
Software engineering
open
7.5%
7.5%
Open-source entry (E. Rocha) — Laude Institute / Konwinski 2025-07
92.5%4.65
High
src
ScienceAgentBench
ML research / Data science
open
33.3%
33.3%
SAB Self-Debug + o3 Medium ($11.69/run) — OSU NLP (HAL) 2025-04
66.7%4.65
High
src
WebGames
Web / Computer-use / GUI
open
48%
48%
Best browser agent (VLM + Browser Use framework) — Convergence AI 2026
52%4.6
High
src
ARC-AGI-3
Frontier reasoning
open
12.6%
12.58%
Tufa Labs 'StochasticGoose' (RL-style exploration agent) — ARC Prize 2026-03
87.4%4.55
High
src
Legal Agent Benchmark (LAB)
Tool-use / Assistant
open
13.3%
13.3%
Claude Fable 5 / Mythos 5 13.3 (held-out, vendor); Opus 4.8 10.4; GPT-5.5 2.1; Gemini 3.1 Pro 0.0 — Harvey 2026-06
86.7%4.5
High
src
TheAgentCompany
Long-horizon / Economic
open
30%
30%
Gemini 2.5 Pro / Claude 3.7 + OpenHands — CMU 2025
70%4.5
High
src
AssistantBench
Tool-use / Assistant
open
25.2%
25.2%
SPA (official); ~42% claimed by aggregators — Allen AI / Tel Aviv 2024
74.8%4.45
High
src
FrontierCode (Diamond)
Software engineering
open
29.3%
29.3%
Claude Fable 5 / Mythos 5 29.3 (xhigh, vendor) — SUSPENDED; independently-published leader Opus 4.8 13.4 — Cognition 2026-06
70.7%4.35
High
src
AutomationBench
Tool-use / Assistant
open
17.4%
17.4%
Claude Fable 5 / Mythos 5 17.4 (vendor, private held-out); Opus 4.8 15.5; GPT-5.5 12.9; Gemini 3.1 Pro 9.6 — Zapier 2026-06
82.6%4.3
High
src
Blueprint-Bench 2
Frontier reasoning
open
38.6%
38.6%
Claude Fable 5 / Mythos 5 38.6 (vendor, contested); GPT-5.5 36.2; Andon's own writeup ranks GPT-5.5 #1 — Andon Labs 2026-06
61.4%4.3
High
src
VisualWebArena
Web / Computer-use / GUI
open
36.5%
36.5%
Recon-Act (GPT-4o-class) — Academic (CMU) 2025
63.5%4.3
High
src
Multi-step cyber autonomy (AISI)
Cyber / Autonomy
active
68.8%
22 steps
Claude Mythos Preview (researcher harness) — first to complete the 32-step range; Opus 4.6 ~16 — UK AISI 2026-04
31.2%4.2
High
src
SWE-Lancer
Software engineering
active
66.3%
66.3%
GPT-5.1 Codex (OpenAI harness) — OpenAI 2026-06
33.7%4.15
High
src
WorkArena++
Web / Computer-use / GUI
open
2.1%
2.1%
Claude Sonnet + BrowserGym (full L2/L3 ~0-2%; easier L2 subset ~39) — ServiceNow Research / academic 2026
97.9%4.15
High
src
FrontierMath
Frontier reasoning
active
52.4%
52.4%
GPT-5.5 Pro (+ code/Epoch scaffold); Fable 5 claimed 87 T1-3 / 88 T4 (suspended, UNVERIFIED) — OpenAI / Epoch AI 2026-05
47.6%4.05
High
src
SWE-bench Multimodal
Software engineering
active
59%
59%
Claude Mythos Preview (self-reported) — Anthropic 2026-06
41%4.05
High
src
SWE-bench Pro
Software engineering
near
80.3%
80.3%
Claude Fable 5 80.3 (Anthropic scaffold, vendor) — SUSPENDED 2026-06-12; SEAL-standardized top GPT-5.4 xHigh + mini-swe-agent 59.1 — Anthropic / Scale SEAL 2026-06
19.7%4.05
High
src
RE-Bench
ML research / Data science
nosota
%

Claude 3.5 Sonnet / o1-preview + METR scaffold — METR 2024-11
%4.0
High
src
tau-bench
Tool-use / Assistant
active
70%
70%
Claude Sonnet 4.5 (airline ~70); overall record Mythos 5 / Fable 5 tied 89.2 (suspended) — Sierra (orig) 2026-06
30%4.0
High
src
Vending-Bench 2
Long-horizon / Economic
open
17.4%
10937 $
Claude Opus 4.7 ~$10,937 (single-player); Opus 4.6 $8,018 official board — Andon Labs 2026
82.6%4.0
High
src
GAIA2 (ARE)
Tool-use / Assistant
nosota
%

Frontier models in ARE harness (Pareto vs cost) — Meta (FAIR) 2025-09
%3.95
High
src
Multi-SWE-bench
Software engineering
nosota
%

n/a — ByteDance Seed 2026
%3.95
High
src
OfficeBench
Office / productivity
open
47%
47%
GPT-4 (best at publication) — UCSD / authors 2024
53%3.9
High
src
ToolSandbox
Tool-use / Assistant
nosota
%

Proprietary >> open (no single SOTA) — Apple 2024-08
%3.9
High
src
Mind2Web 2
Web / Computer-use / GUI
active
60%
60%
OpenAI Deep Research (o-series agent) — OpenAI 2025-06
40%3.85
High
src
GAIA
Tool-use / Assistant
active
74.6%
74.6%
HAL Generalist Agent + Claude Sonnet 4.5 (val, reproducible single-agent) — Princeton HAL / orig Meta-HF 2026-05
25.4%3.8
High
src
Humanity's Last Exam (HLE)
Frontier reasoning
active
64.7%
64.7%
Claude Mythos Preview 64.7 (tools) / Fable 5 64.5 (tools), 59.0 (no-tools) — SUSPENDED; AA text-only lists Fable 5 53 — Anthropic 2026-06
35.3%3.8
High
src
tau2-bench
Tool-use / Assistant
near
76.5%
76.5%
LongCat-Flash-Thinking-2601 (airline 76.5); telecom ~99 (JT-35B-Flash 99.1); Fable 5 telecom 98.5 — Sierra (orig) 2026-06
23.5%3.8
High
src
FutureX
Deep research
active
56%
56%
H2O AI Super Agent — H2O.ai (board by FutureX team) 2026
44%3.75
High
src
APEX-Agents-AA
Tool-use / Assistant
open
47.1%
47.1%
Gemini 3.5 Flash (high) on AA Stirrup harness — Artificial Analysis (data by Mercor) 2026-05
52.9%3.7
High
src
BFCL v4
Tool-use / Assistant
active
73%
73%
Qwen3.7 Max (native FC) — UC Berkeley (Gorilla) 2026-06
27%3.7
High
src
PaperBench
ML research / Data science
open
24.4%
24.4%
orig o1-high 24.4 (verified); Kimi K2.5 63.5 self-reported/unverified — Moonshot AI / orig OpenAI 2026-05
75.6%3.7
High
src
GTA / GTA-2
Tool-use / Assistant
nosota
%

Frontier models in GTA harness (GPT-4 era <50) — OpenCompass / Shanghai AI Lab 2026-02
%3.65
High
src
SWE-rebench
Software engineering
active
65.3%
65.3%
Claude Opus 4.6 + fixed 128K ReAct scaffold — Nebius 2026-05
34.7%3.6
High
src
AgentBench
Tool-use / Assistant
nosota
%

Frontier models (general); no verified 2026 SOTA — Tsinghua (THUDM)
%3.55
Med
src
ARC-AGI-2
Frontier reasoning
active
52.9%
52.9%
GPT-5.2 X-High ~52.9% @ $1.90/task (efficiency-limited; first to ~match human) — OpenAI / Google 2026-05
47.1%3.55
Med
src
ExploitBench
Cyber / Autonomy
near
78%
78%
Claude Mythos 5 (unblocked, restricted) 78.0; Mythos Preview 69.0; Opus 4.8 40.0; GPT-5.5 34.0 — CMU + Bugcrowd 2026-06
22%3.55
Med
src
LiveBench (agentic-coding subset)
Software engineering
nosota
%

not separately reported — LiveBench (Abacus.AI / NYU / Nvidia) 2025-04
%3.55
Med
src
BioMysteryBench
ML research / Data science
open
46.1%
46.1%
Claude Fable 5 / Mythos 5 46.1 hard / 83.9 human-solvable (vendor); Opus 4.8 40.0 / 80.4 — Anthropic 2026-06
53.9%3.5
Med
src
DeepResearch Bench (RACE)
Deep research
active
56%
55.95 score
NVIDIA AI-Q deep research agent — NVIDIA 2026-03
44%3.4
Med
src
GDPval
Long-horizon / Economic
open
41%
41%
Board now Elo: Claude Fable 5 GDPval-AA Elo 1932 (#1, SUSPENDED); Opus 4.8 1890; GPT-5.5 1769 — OpenAI / Artificial Analysis 2026-06
59%3.4
Med
src
NYU CTF Bench
Cyber / Autonomy
active
59%
59%
Claude Opus 4.5 (CTF agent harness) — NYU 2026-04
41%3.4
Med
src
WebArena
Web / Computer-use / GUI
active
71.6%
71.6%
OpAgent v2 (VLM + Planner/Grounder/Reflector, online RL) — academic (arXiv 2602.13559) 2026-04
28.4%3.4
Med
src
SpreadsheetBench
Office / productivity
active
70.5%
70.5%
Gemini in Google Sheets (vendor) — Google 2026-03
29.5%3.35
Med
src
Aider Polyglot
Software engineering
near
88%
88%
gpt-5 high (88%) / Opus 4.5 (89.4%, vendor) via aider — OpenAI / Anthropic 2025-08
12%3.3
Med
src
DSBench
ML research / Data science
open
34%
34%
DatawiseAgent (notebook-centric); modeling >90 — Academic 2025-03
66%3.25
Med
src
MLE-bench
ML research / Data science
active
64.4%
64.4%
Famou-Agent 2.0 + Gemini-3-Pro (24h/task) — Baidu 2026-02
35.6%3.25
Med
src
BIRD-SQL
Data / SQL
near
81.7%
81.67%
Agentar-Scale-SQL (test-time scaling) — Ant Group / community 2026
18.3%3.2
Med
src
CORE-Bench (Hard)
ML research / Data science
near
77.8%
77.8%
Claude Code + Claude Opus 4.5 ($87/run) — Princeton HAL 2025-12
22.2%3.2
Med
src
HealthBench Professional
Tool-use / Assistant
active
66%
66%
Claude Fable 5 / Mythos 5 66.0 (vendor); Opus 4.8 56.9; GPT-5.5 51.8; OpenAI GPT-5.4 'ChatGPT for Clinicians' beats physician baseline — OpenAI 2026-06
34%3.2
Med
src
Spider 2.0
Data / SQL
active
73.1%
73.13%
DivSkill-SQL (Lite 73.13); Genloop Sentinel Agent v2 Pro (full-set 96.7) — Snowflake/UCSD; Genloop 2026-06
26.9%3.2
Med
src
WebDev / Code Arena (LMArena)
Web / Computer-use / GUI
nosota
%
1567 Elo
Claude Opus 4.7 Thinking — LMArena 2026-05
%3.2
Med
src
Terminal-Bench 2.0
Software engineering
near
84.7%
84.7%
NexAU-AHE harness + GPT-5.5 (official board) — OpenAI model / NexAU 2026-05
15.3%3.15
Med
src
Terminal-Bench 2.1
Software engineering
near
83.4%
83.4%
Codex CLI (GPT-5.5) 83.4 (public board); Anthropic claims Fable 5 / Mythos 5 88.0* (vendor, not on board, suspended) — Stanford/Laude (Harbor) 2026-06
16.6%3.15
Med
src
ScreenSpot-Pro
Web / Computer-use / GUI
near
87.9%
87.9%
Claude Opus 4.8 (native grounding, general); specialist GTA1-32B 95.2 — Anthropic 2026-06
12.1%3.05
Med
src
AndroidWorld
Web / Computer-use / GUI
near
87.1%
87.1%
H Company Surfer 2 (vision-only, vendor 87.1); verified board top Qwen3.5-35B-A3B 71.1 (model-only) — H Company / Alibaba 2026
12.9%3.0
Med
src
BrowseComp
Web / Computer-use / GUI
saturated
90.1%
90.1%
GPT-5.5 Pro (agentic deep-research scaffold) — OpenAI 2026-06
9.9%3.0
Med
src
SWE-bench Verified
Software engineering
saturated
95%
95%
Claude Fable 5 95.0 (Anthropic harness) — SUSPENDED 2026-06-12; Mythos 5 95.5 (restricted); GA-usable GPT-5.5/Opus 4.8 ~88.7 — Anthropic 2026-06
5%2.9
Low
src
OSWorld (Verified)
Web / Computer-use / GUI
near
85.4%
85.4%
Claude Mythos Preview 85.4 (Anthropic system card); Fable 5 85.0; Opus 4.8 83.4 (GA-usable); GPT-5.5 78.7 — Anthropic 2026-06
14.6%2.75
Low
src
Online-Mind2Web
Web / Computer-use / GUI
saturated
97%
97%
Browser Use Cloud v3 (Auto-Research scaffold) — Browser Use 2026-03
3%2.55
Low
src
Cybench
Cyber / Autonomy
saturated
100%
100%
Claude Mythos Preview ~100 (self-reported) — Anthropic (eval: Stanford) 2026-06
0%2.1
Low
src
WebVoyager
Web / Computer-use / GUI
saturated
98.5%
98.5%
Alumnium (Claude Sonnet 4.6 policy + GPT-5 Nano) + Selenium — Alumnium 2026-03
1.5%1.9
Low
src
AgentHarm
Autonomy / Safety
nosota
%

n/a (safety eval; inverse metric) — Gray Swan + UK AISI 2024-10
%
n/a (safety eval)
src
Artificial Analysis Intelligence Index
Meta-leaderboards
tracker
%

Claude Fable 5 64.9 (#1, SUSPENDED 06-12); GPT-5.5 ~59.9 — Artificial Analysis 2026-06
%
n/a (tracker)
src
Epoch AI Benchmarking Hub
Meta-leaderboards
tracker
%

ECI 159: GPT-5.5 Pro (Fable 5 not yet recomputed) — Epoch AI 2026-04
%
n/a (tracker)
src
Galileo Agent Leaderboard v2
Meta-leaderboards
tracker
%

n/a — Galileo AI
%
n/a (tracker)
src
HAL - Holistic Agent Leaderboard
Meta-leaderboards
tracker
%

n/a — Princeton PLI
%
n/a (tracker)
src
METR Time Horizon
Autonomy / Safety
nosota
%
14.5 hours
Claude Opus 4.6 (>16h Mythos preview, unreliable) — METR 2026-02
%
n/a (measurement)
src
Vellum LLM Leaderboard
Meta-leaderboards
tracker
%

n/a — Vellum
%
n/a (tracker)
src

agent = Model + Harness · snapshot 2026-06 · verify numbers against each benchmark's primary source.