Fireworks

GLM 5.3

Made by Z.ai (Zhipu AI). Released 2026-08-14. Accessibility not recorded. Cheapest standard rate we hold: $4.40 per million output tokens, no context split.

What it costs

TierKindInOutCached inFrom
standardtext$1.40$4.40$0.26*registry-single

per million tokens, US dollars. * marks a rate from a single registry or two that disagree — the best available number rather than a confirmed one. Every other row is confirmed against a seller or a licensed index.

What it scores, by house

BenchmarkHouseScoreEvaluatedScaffold
chess puzzlesEpoch AI0.2±0.042026-08-24max
frontiermath tier 4 v2Epoch AI0.3±0.072026-08-25max
frontiermath tiers 1 3 v2Epoch AI0.7±0.032026-08-25max
gpqa diamondEpoch AI0.9±0.022026-08-24max
mystery game puzzlesEpoch AI0.3±0.052026-08-30max
otis mock aime 2024 2025near its ceilingEpoch AI0.9±0.032026-08-24max
simpleqa verifiedEpoch AI0.4±0.022026-08-28max
ale benchcollected by Epoch AI1317high
critptcollected by Epoch AI0.2max
deepswecollected by Epoch AI0.7mini-swe-agent
frontierswecollected by Epoch AI0.3proximus
proofbenchcollected by Epoch AI0.5max
scicodecollected by Epoch AI0.6max
vending bench 2collected by Epoch AI8164unknown
webdev arenacollected by Epoch AI1599max
weirdmlcollected by Epoch AI0.8max

These scores are not comparable between houses and are not added up. A score is dated by when it was evaluated, not when the model was released, and a score without its scaffold is not a measurement — which is why both are printed.

Every rate we hold · Why there is no single best model · Everything from Fireworks