Electricity Bench

Claude Sonnet 5.5 vs GPT-6 Astra

GPT-6 Astra leads overall, C to C-. Claude Sonnet 5.5 is about 18x cheaper per task. By suite, Claude Sonnet 5.5 takes real-world issues (C- to D) and spec planning is even at C+. Take Claude Sonnet 5.5 anyway when you want the side that is cheaper per task, faster on real-world issues and ahead on real-world issues.Measured

Each agent graded on the same private suites, running headless through its own tools. Advantages are oriented so a win is a win, whichever way the metric runs.

Verdict: GPT-6 Astra vs Claude Sonnet 5.5

GPT-6 Astra leads overall, C to C-. By suite: Claude Sonnet 5.5 takes real-world issues, C- to D (60% vs 43% capability); spec planning is even at C+ (74% vs 74% capability); GPT-6 Astra takes vibe coding, C+ to D+ (74% vs 48% capability). Claude Sonnet 5.5 used <1% of Claude Max 5x's weekly limit per pass, $0.003 per task at the listed price. GPT-6 Astra used 22% of ChatGPT Plus's weekly limit per pass, $0.053 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~4.4% of ChatGPT Pro 5x's weekly limit (est, about $0.053 per task). Per task, Claude Sonnet 5.5 is about 18x cheaper ($0.003 vs $0.053). Claude Sonnet 5.5 is faster on real-world issues: a median task takes 37s against 2m 13s. Both finished every task on it. Reasons to pick Claude Sonnet 5.5 anyway: it is cheaper per task ($0.003 vs $0.053), it is faster on real-world issues (37s vs 2m 13s median) and it leads on real-world issues (C- to D).

Claude Sonnet 5.5Clears tier 4 (partial tier 5) at 4.0% of Claude Max 5x per runC-OVERALLGPT-6 AstraClears tier 2 (partial tier 5) at 17.0% of ChatGPT Plus per runCOVERALL
real-world@v3C-vsD
DimensionClaude Sonnet 5.5GPT-6 AstraAdvantage
Capability60%43%+17% (Claude Sonnet 5.5 leads)
Median task time37s2m 13s1m 36s faster (Claude Sonnet 5.5 leads)
Approx. task cost$0.002 / task$0.052 / task$0.050 / task lower (Claude Sonnet 5.5 leads)

Claude Sonnet 5.5: 4.0% of Claude Max 5x (4.0% / 5h) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. GPT-6 Astra: 17.0% of ChatGPT Plus (112.0% / 5h, 17.0% / 1w) per run. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.

spec-planning@v1C+vsC+
DimensionClaude Sonnet 5.5GPT-6 AstraAdvantage
Capability74%74%+0% (Claude Sonnet 5.5 leads)
Median task time3m 25s3m 5s19s faster (GPT-6 Astra leads)
Approx. task cost$0.007 / task$0.057 / task$0.051 / task lower (Claude Sonnet 5.5 leads)

Claude Sonnet 5.5: 4.0% of Claude Max 5x (4.0% / 5h) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. GPT-6 Astra: 5.0% of ChatGPT Plus (27.0% / 5h, 5.0% / 1w) per run. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.

vibe-coding@v1D+vsC+
DimensionClaude Sonnet 5.5GPT-6 AstraAdvantage
Capability48%74%+26% (GPT-6 Astra leads)
Median task time56s2m 40s1m 44s faster (Claude Sonnet 5.5 leads)
Approx. task cost$0.001 / task$0.046 / task$0.045 / task lower (Claude Sonnet 5.5 leads)

Claude Sonnet 5.5: 0.2% of Claude Max 5x (0.2% / 5h) per good solve. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. GPT-6 Astra: 1.0% of ChatGPT Plus (7.2% / 5h, 1.0% / 1w) per good solve. Its weekly limit holds about 6 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.

Grades and deltas are comparable only within a suite version. See methodology.

Frequently asked questions

Which is better for coding, Claude Sonnet 5.5 or GPT-6 Astra?

Claude Sonnet 5.5 clears higher: tier 4 of 5 versus tier 2 for GPT-6 Astra on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-09-28.

Which is cheaper per task, Claude Sonnet 5.5 or GPT-6 Astra?

Claude Sonnet 5.5 used <1% of Claude Max 5x's weekly limit per pass, $0.003 per task at the listed price. GPT-6 Astra used 22% of ChatGPT Plus's weekly limit per pass, $0.053 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~4.4% of ChatGPT Pro 5x's weekly limit (est, about $0.053 per task). Per task, Claude Sonnet 5.5 is about 18x cheaper ($0.003 vs $0.053).

What do Claude Sonnet 5.5 and GPT-6 Astra cost a month?

Claude Sonnet 5.5 runs on Claude Max 5x at $100/month (public price as of 2026-07-20). GPT-6 Astra runs on ChatGPT Plus at $20/month (public price as of 2026-07-20).