pgai.MODEST PRAISE. PUBLIC NUMBERS.

THE AI PERFORMANCE DEPARTMENT

Pretty great.
At what, exactly?

A little less hype. A few more receipts.
AI models, judged one benchmark at a time.

Data as of 2026-10-14 / manually checked

4 models · 7 benchmarks · zero composite scores

pretty greatpretty okaynot yetTap a verdict for the number and source.
MODEL / VENDORSWE-bench VerifiedcodingTerminal-Bench 2.0terminal workARC-AGI-2abstract reasoningGPQA DiamondscienceAIME 2025mathLMArena / Text ArenaconversationMRCRv2 · 128k–256klong context
OpenAI

GPT-5.2 Thinking

Pretty great at coding, abstract reasoning, science, math, long context.

SWE-bench Verified
pretty great
80%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
Terminal-Bench 2.0 not reported
ARC-AGI-2
pretty great
52.9%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
GPQA Diamond
pretty great
92.4%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
AIME 2025
pretty great
100%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
LMArena / Text Arena not reported
MRCRv2 · 128k–256k
pretty great
77%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
OpenAI

GPT-5.1 Thinking

Pretty great at coding, science.

SWE-bench Verified
pretty great
76.3%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
Terminal-Bench 2.0 not reported
ARC-AGI-2
not yet
17.6%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
GPQA Diamond
pretty great
88.1%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
AIME 2025
pretty okay
94%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
LMArena / Text Arena not reported
MRCRv2 · 128k–256k
not yet
29.6%

Vendor-reported research evaluation, maximum available reasoning effort. See source for harness and sampling.

Source ↗Observed 2026-10-14
Anthropic

Claude Opus 4.6

Pretty great at coding, terminal work.

SWE-bench Verified
pretty great
80.8%

Vendor-reported SWE-bench Verified. See source for evaluation and memorization-screen caveats; harness differs from OpenAI reports.

Source ↗Observed 2026-10-14
Terminal-Bench 2.0
pretty great
65.4%

Vendor-reported. Terminus-2 harness, adaptive thinking at maximum effort, 1 million tokens per task; 1× guaranteed / 3× ceiling resources, five attempts per task.

Source ↗Observed 2026-10-14
ARC-AGI-2 not reported
GPQA Diamond not reported
AIME 2025 not reported
LMArena / Text Arena
pretty okay
1498 Elo

Text Arena overall, no adjustments. Leaderboard dated October 13, 2026. ±3.

Source ↗Observed 2026-10-14
MRCRv2 · 128k–256k not reported

The top score means the top sourced score in this snapshot. Different harnesses, budgets and samples can change results. A blank cell is a blank cell.