Code SmithArchitectPeer public scoresmeth_v0_1_1as of 2026-07-11

Claude Opus 4.8 · API · High effort (default)

The expensive architect — pays for hard repos, long agents, and fewer retries.

Flagship coding ceiling when constraint-heavy, multi-hour, or autonomous work justifies $5/$25.

SharePost on X

Codex rating

66.9

Confidence B

Overall mean of attribute public scores after peer min-max (40–95 band within scored launch peer set).

Loadout

Endpoint: claude-opus-4-8
Product: Anthropic API
Price: $5 / $25 per MTok
Context: 1000k tokens

Flagship Opus 4.8 API, high effort, 1M context, $5/$25 — expensive coding ceiling.

Core attributes

Six stats · peer public

Peer publicbars = rank in scored peer set (≈40–95) · raw = absolute fusion under the bar

STRExecution

77.5raw 83.7A73.5–81.5

DEXTempo

47.3raw 52.0B37.3–57.3

CONReliability

76.0raw 83.5A72–80

INTReasoning

84.5raw 86.6A80.5–88.5

WISJudgment

75.7raw 72.3A71.7–79.7

CHACollaboration

40.1raw 72.9A36.1–44.1

Peer public (main number / bar) ranks this Build inside the current scored peer set (≈40–95). Raw is the absolute evidence-fusion aggregate before min-max. A low peer score can still be a strong absolute model — see methodology.

Practical verdict

When to choose

Best at

Hard repository migration and multi-file refactors
Autonomous coding agents with named harnesses
High-stakes reasoning and verification loops

Avoid for

Everyday chat and low-stakes drafts (use Sonnet/Flash)
Strict budget / high-volume token farms
Latency-first interactive sprinters

Choose this when

The job is hard enough that a wrong answer costs more than Opus tokens.

Skills

Task proficiencies

Creative Writing90.0 B

Professional Writing90.0 B

Test Writing88.6 A

Devops And Terminal84.6 A

Tool Calling84.6 A

Refactoring84.4 A

Debugging83.6 A

Existing Codebase Work82.4 A

Quantitative Reasoning81.9 A

Long Autonomous Runs81.1 A

Research Synthesis79.1 A

Following Complex Instructions67.3 A

Fast Interactive Work52.0 B

Traits & Quirks

Signature strengths and honest flaws

No structured traits or flaws catalogued for this Build yet. Traits are behavioural patterns (beyond raw scores) — added as evidence and field observation accumulate. See Best at / Avoid for above in the meantime.

Evidence Terminal

Traceability

Every factual claim cites an evidence row. Tap a chip to see the benchmark, exact config, source, and caveats behind the number.

GPQA Diamond92.02%A Humanity's Last Exam45.74%A AA-Omniscience27.43A Artificial Analysis API latency / throughput62B Chatbot Arena / LMArena human preference1506B Humanity's Last Exam57.9%B IFBench62.24%A SWE-bench Pro69.2%B SWE-bench Verified88.6%A Terminal-Bench84.64%A

Engine pass meth_v0_1_1 (peer min-max vs scored launch peer set). Codex overall 66.9 (B). Attribute public scores map raw aggregates into a 40–95 band within the peer set (methodology min-max). INT uses hierarchical fusion (GPQA primary / HLE secondary / capped bleed).