Claude Opus 5 vs GLM-5.3

head-to-head

what do the benchmarks actually say?

On SWE-bench Verified — real, human-validated GitHub issues resolved end-to-end — Claude Opus 5 posts 97.0% against 95.4% for GLM-5.3, a 1.6-point gap. Verified is the closest public proxy for "can it fix a real bug in a real repo without help", which is why it anchors our ranking.

A few points either way is real but not decisive: within that band, the agent scaffolding around the model — how it retrieves files, runs tests, and retries — often matters as much as the base model. Treat the gap as a lean, not a verdict.

which is cheaper to run?

Public per-token pricing isn't confirmed for both models, so we don't print a cost comparison yet.

when to pick each

how were these scores verified?

We only print a number once it's confirmed against a primary source or an independent evaluation, and each row on our leaderboard records which kind it is:

• Claude Opus 5: Independent (vals.ai, observed Jul 25 2026, mini-swe-agent bash-only harness): SWE-bench Verified 97.00% ±0.76, the highest score on the board and 1st of the 75 systems vals.ai has run. Entered ranked Jul 25, 2026 after one day unranked: Anthropic published no SWE-bench Verified number at launch and still has not, so this is vals.ai's own measurement rather than a vendor claim. Read the #1 as a three-way tie, not a win — GPT-5.6 Sol is at 96.20% ±0.86 (a 0.8-point gap, ~0.7 sigma) and Claude Fable 5 at 95.00% ±0.98 (2.0 points, ~1.6 sigma), both inside the combined margin of error. We rank Opus 5 first only because it scored highest. Where the top two genuinely separate is task length, and not in Opus 5's favour: on the 1-to-4-hour tier Sol solves 98% against Opus 5's 90%, while Opus 5 leads on shorter work (98% under 15 minutes and 97% on 15-minute-to-1-hour tasks, vs 97% and 95% for Sol). Released Jul 24, 2026 at $5/$25 per 1M, the same price as Opus 4.8 and half of Fable 5. Anthropic's launch claims stay unreproducible (Frontier-Bench v0.1, CursorBench 3.2 and Zapier AutomationBench are proprietary), so this is the first externally checkable score the model has. Fast mode runs about 2.5x default speed at 2x base price.
• GLM-5.3: Independent (vals.ai, 2026-08-20 sweep, mini-swe-agent bash-only harness): SWE-bench Verified 95.4% ±0.94, 6th of 86 systems. Left the verifying queue on 2026-08-20, six days after launch. Z.ai published NO SWE-bench Verified figure of its own, so this rank rests entirely on the independent run, which is the cleanest kind of row on this board. Read it as tied with GPT-5.6 Terra (95.4% ±0.94), Grok 4.6 (95.6%) and Claude Fable 5 (95.0%); pooled SEM is about 1.3 points and every gap is inside it. That result is striking given the release is post-training only, on the same base model as GLM 5.2, which is ranked far below on an independent 82.8%. Vendor numbers from the Aug 14 2026 release post, run mostly inside a Claude Code 2.1.207 harness at max effort rather than a neutral one: Terminal-Bench 3.0 28.3 (up from GLM-5.2’s 4.6), DeepSWE v1.1 66.9 (from 46.2), Agents’ Last Exam 28.5 (from 23.8). Its headline claims were cyber rather than coding: CyberGym 84.5% and ExploitBench 54.4%. Marked closed because the weights are NOT out: Z.ai said roughly two weeks after launch pending safety hardening, with no license announced, so the open-weights flag flips only when they actually land. See /p/glm-5-3-cybergym-open-weights-delayed/.

Full reviewsClaude Opus 5, decodedGLM-5.3, decoded

Ranked on our AI Coding Leaderboard, updated 2026-08-20. Scores are confirmed against primary sources; prices are per 1M input tokens and can change.