DeepInfra

GLM-5

Made by Z.ai (Zhipu AI). Released 2026-02-11. Accessibility not recorded. Cheapest standard rate we hold: $2.08 per million output tokens, no context split.

What it costs

TierKindInOutCached inFrom
standardtext$0.60$2.08$0.12registry

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 AI0.1±0.032026-02-12
frontiermathEpoch AI0.2±0.022026-02-19
frontiermath tier 4Epoch AI0.0±0.022026-02-19
gpqa diamondEpoch AI0.9±0.022026-02-12
otis mock aime 2024 2025near its ceilingEpoch AI0.8±0.062026-02-12
swe bench verifiedEpoch AI0.7±0.022026-02-15
ale benchcollected by Epoch AI766
apex agentscollected by Epoch AI0.3
arc aginear its ceilingcollected by Epoch AI0.4
arc agi 2collected by Epoch AI0.0
cl benchcollected by Epoch AI0.2
forecastbenchcollected by Epoch AI61.0
simplebenchcollected by Epoch AI0.5
terminalbenchcollected by Epoch AI0.5Terminus 2
vending bench 2collected by Epoch AI4432
webdev arenacollected by Epoch AI1435
weirdmlcollected by Epoch AI0.5

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 DeepInfra