Skip to content
Pareto Inference

Every provider of GLM 5.3 Flash from Z.ai that OpenRouter, Vercel AI Gateway and models.dev list, cheapest first: input, output and cached-input prices per 1M tokens, context, speed and uptime. Checked every hour.

Pareto Inference runs this page and is one of the providers listed. Rows are sorted by price, and Pareto's rows are labeled and placed by price like any other. The methodology says what is included and what is left out. Spot a wrong number? Email us.

Cheapest right now

USD per 1M tokens at 17:32 UTC, peak hours applied.

Agent blend$0.01017Pareto Inference, Vercel AI Gateway
3:1 blend$0.04445Relace (fp4), OpenRouter or Vercel AI Gateway
Input$0.0285Relace (fp4), OpenRouter or Vercel AI Gateway
Cached input$0.0075Pareto Inference, Vercel AI Gateway
Output$0.0923Relace (fp4), OpenRouter or Vercel AI Gateway

Providers

Sorted by the agent blend now: the price of 1M tokens that are 5% uncached input, 94% cached input and 1% output, the mix measured on Pareto's coding-agent traffic. Under it, the 3:1 blend: three input tokens to one output token, cache ignored. Speed and time to first token are Vercel AI Gateway's medians for the last hour; uptime is the last day's, from Vercel AI Gateway where it lists the provider, else OpenRouter.

104 offers from 93 providers · USD per 1M tokens
ProviderBlendedInputOutputCached inputContextSpeedUptime
Pareto InferenceVercel AI GatewayRuns this page$0.010173:1 $0.05937$0.0375$0.125$0.00751M131K out63 tok/s2.34 s TTFT100%
Pareto InferenceDirectRuns this pagePromo, 33% off$0.016283:1 $0.095$0.06$0.20$0.0121M131K out——
Darkbloomfp4OpenRouter$0.016333:1 $0.09375$0.065$0.18$0.012262K65.5K out—95.71%
Near AIfp8OpenRouter$0.020283:1 $0.185$0.08$0.50$0.0121.05M944K out—100%
Standard ComputeDirect, per models.dev$0.020283:1 $0.185$0.08$0.50$0.0121M50K out——
302.AIDirect, per models.dev$0.020353:1 $0.1187$0.075$0.25$0.0151M131K out——
DeepInfraVercel AI Gateway$0.020353:1 $0.1187$0.075$0.25$0.0151.05M1.05M out129 tok/s1.37 s TTFT99.99%
GMICloudfp8OpenRouter · Vercel AI GatewayPromo, 50% off$0.020353:1 $0.1187$0.075$0.25$0.0151.05M944K out68 tok/s0.75 s TTFT98.68%
Tenstorrentfp8OpenRouter$0.020353:1 $0.1187$0.075$0.25$0.015262K131K out—97.98%
ZenMuxDirect, per models.dev$0.020353:1 $0.1187$0.075$0.25$0.0151M128K out——
Novitafp8OpenRouterPromo, 44% off$0.022793:1 $0.133$0.084$0.28$0.01681.05M131K out—99.24%
StreamLakefp8OpenRouterPromo, 41% off$0.024013:1 $0.1401$0.0885$0.295$0.01771.02M128K out—97.71%
NovitaVercel AI Gateway$0.024423:1 $0.1425$0.09$0.30$0.0181M128K out42 tok/s2.16 s TTFT100%
OpenInferencefp4OpenRouter$0.024583:1 $0.0825$0.03$0.24$0.0221.05M944K out—100%
CortecsDirect, per models.dev$0.025423:1 $0.1625$0.10$0.35$0.0181.05M1.05M out——
Sail ResearchVercel AI Gateway$0.02783:1 $0.17$0.11$0.35$0.021.05M131K out70 tok/s0.42 s TTFT100%
IteraComputeDirect, per models.dev$0.028423:1 $0.13$0.08$0.28$0.0231.05M131K out——
CoreWeavenvfp4OpenRouter$0.02863:1 $0.185$0.12$0.38$0.021.05M944K out—100%
Relacefp4OpenRouter · Vercel AI Gateway$0.029143:1 $0.04445$0.0285$0.0923$0.02851.05M131K out77 tok/s1.37 s TTFT99.85%
DevPass (LLM Gateway)Direct, per models.dev$0.03043:1 $0.1285$0.088$0.25$0.0251.05M131K out——
Phalanvfp4OpenRouterPromo, 25% off$0.030533:1 $0.1781$0.1125$0.375$0.02251.05M131K out—99.67%
RunInfraVercel AI Gateway$0.030563:1 $0.156$0.088$0.36$0.0241.05M1.05M out296 tok/s2.04 s TTFT100%
KenariDirect, per models.dev$0.030633:1 $0.1714$0.1143$0.3429$0.022861M131K out——
MeliousDirect, per models.dev$0.0313:1 $0.1951$0.1115$0.446$0.02231M131K out——
NanoGPTDirect, per models.dev$0.03153:1 $0.15$0.10$0.30$0.0251.05M131K out——
BoundlessVercel AI Gateway$0.032563:1 $0.19$0.12$0.40$0.0241.05M1.05M out120 tok/s1.91 s TTFT100%
VancineDirect, per models.dev$0.032563:1 $0.19$0.12$0.40$0.0241M131K out——
GreenPTDirect, per models.dev$0.035523:1 $0.2236$0.1278$0.511$0.025551M131K out——
WaferusOpenRouter$0.035853:1 $0.07247$0.0353$0.184$0.03431.05M944K out—100%
AIHubMixDirect, per models.dev$0.036063:1 $0.1831$0.1127$0.3944$0.028171M131K out——
Sail Researchus, fp4OpenRouter$0.036153:1 $0.2875$0.175$0.625$0.02251.05M131K out—99.42%
Munito AIDirect, per models.dev$0.036383:1 $0.21$0.13$0.45$0.0271.05M944K out——
engyDirect, per models.dev$0.036633:1 $0.2137$0.135$0.45$0.027328K32.8K out——
TelnyxVercel AI Gateway$0.036633:1 $0.2137$0.135$0.45$0.0271.05M1.05M out270 tok/s0.35 s TTFT100%
RekaOpenRouter$0.03723:1 $0.165$0.12$0.30$0.031.05M944K out—100%
Particle AIVercel AI Gateway$0.03723:1 $0.175$0.10$0.40$0.031M131K out35 tok/s0.84 s TTFT99.99%
DigitalOceanOpenRouter$0.037573:1 $0.1781$0.1125$0.375$0.031.05M131K out—100%
Makorafp4OpenRouter$0.03763:1 $0.165$0.14$0.24$0.031.05M131K out—99.79%
InferenceNetVercel AI Gateway$0.03773:1 $0.1875$0.10$0.45$0.031.05M128K out131 tok/s0.37 s TTFT100%
Particle AInvfp4OpenRouter$0.03823:1 $0.20$0.10$0.50$0.031.05M131K out—99.93%
NextBitfp4OpenRouter$0.03873:1 $0.2025$0.12$0.45$0.031.05M128K out—100%
InferenceNetus, fp4OpenRouter$0.038983:1 $0.1685$0.108$0.35$0.0321.05M262K out—100%
AI/ML APIDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out——
ai&Direct, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out——
ArceeDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
AtlasCloudfp8OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out—100%
BaseTenfp8 us, fp8OpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out118 tok/s0.55 s TTFT99.97%
Cloudflare Workers AIDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M1.05M out——
Coherenvfp4OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out—100%
CommandCodeDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
CrossModelDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M128K out——
Crusoefp4OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out—100%
DeepInfrafp4OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out—100%
DigitalOceanVercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M1.05M out144 tok/s0.79 s TTFT99.96%
Eden AIDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out——
FireworksOpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out168 tok/s1.04 s TTFT99.98%
FriendliOpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out119 tok/s0.64 s TTFT100%
Kilo GatewayDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out——
Merge GatewayDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
MistralOpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out—100%
Modalfp4OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out—100%
Munito AI (EU)Direct, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out——
Nebiusfp4OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out—99.61%
NeonDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out——
NeuralwattDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M1.05M out——
OfoxDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
Ollama CloudDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
OpenCodeDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
OrcaRouterDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M128K out——
Parasailfp4OpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out128 tok/s1.33 s TTFT99.99%
Perplexity AgentDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
PrimeIntellectOpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out—99.79%
RunwareVercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.03131K131K out324 tok/s0.33 s TTFT99.96%
SiliconFlowfp8OpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M262K out—99.85%
StreamLakeVercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.02M128K out50 tok/s1.04 s TTFT99.86%
Tempr GatewayDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
Tencent CloudVercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031M128K out46 tok/s2.10 s TTFT100%
TogetherOpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M944K out209 tok/s0.78 s TTFT99.97%
TokenGoDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031M131K out——
Umans AIDirect, per models.dev$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out——
VeniceOpenRouter$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out—99.59%
Z.AIfp8OpenRouter · Vercel AI Gateway$0.04073:1 $0.2375$0.15$0.50$0.031.05M131K out50 tok/s3.90 s TTFT99.56%
Volcengine ArkDirect, per models.dev$0.042193:1 $0.193$0.1187$0.4156$0.034141M131K out——
Zhipu AIDirect, per models.dev$0.042463:1 $0.1942$0.1195$0.4183$0.034361M131K out——
SiliconFlow (China)Direct, per models.dev$0.042613:1 $0.1949$0.1199$0.4198$0.034491.05M131K out——
Charm HyperDirect, per models.dev$0.043293:1 $0.2586$0.1633$0.5444$0.031571.05M131K out——
above.devDirect, per models.dev$0.043743:1 $0.2612$0.165$0.55$0.03191M131K out——
SyntheticDirect, per models.dev$0.05013:1 $0.2375$0.15$0.50$0.04524K65.5K out——
VivgridDirect, per models.dev$0.05013:1 $0.2375$0.15$0.50$0.041M131K out——
Inceptronfp8OpenRouter$0.050853:1 $0.2437$0.175$0.45$0.041.05M944K out—100%
ParasailfastOpenRouter$0.050873:1 $0.2969$0.1875$0.625$0.03751.05M944K out—99.84%
Wafernvfp4OpenRouter · Vercel AI Gateway$0.054523:1 $0.0905$0.054$0.20$0.0531.05M944K out72 tok/s1.07 s TTFT100%
Morphfp8OpenRouter · Vercel AI Gateway$0.057463:1 $0.1535$0.122$0.248$0.0521.05M944K out117 tok/s0.33 s TTFT100%
TensorXDirect, per models.dev$0.0623:1 $0.275$0.20$0.50$0.051.05M64K out——
Privatemode AIDirect, per models.dev$0.07343:1 $0.3611$0.2311$0.7511$0.05781M131K out——
EmpirioLabs AIDirect, per models.dev$0.076753:1 $0.1187$0.075$0.25$0.0751.05M131K out——
OpperDirect, per models.dev$0.08083:1 $0.275$0.20$0.50$0.071M128K out——
RequestyDirect, per models.dev$0.08183:1 $0.30$0.20$0.60$0.071M262K out——
WallyDirect, per models.dev$0.091543:1 $0.3665$0.272$0.65$0.0761.05M131K out——
VultrDirect, per models.dev$0.10253:1 $0.1625$0.10$0.35—1.05M131K out——
ModalVercel AI Gateway$0.12213:1 $0.7125$0.45$1.50$0.091M1M out169 tok/s0.38 s TTFT100%
Hugging FaceDirect, per models.dev$0.15353:1 $0.2375$0.15$0.50—1.05M131K out——
BothubDirect, per models.dev$0.16033:1 $0.248$0.1566$0.522—1M131K out——
Berget.AIDirect, per models.dev$0.29293:1 $0.3625$0.29$0.58—1.05M16.4K out——

Index