Wirestack
FeedBrowseAPI workflowsUpdatesComparePricing
© 2026 WirestackPricingCompareUpdatesUse in your editor

Compare tools

Hard facts side by side: free tiers, prices, limits, SDKs and privacy. Every value links to the official page it came from and is re-checked every week.

Start from what you need:ChatVoiceVisionImage genVideo genSearchEmbeddingsAgentsTool callingAutomationTranslationSummaries
FactGroqMistral AIOpenRouter
Free tierYes

Free tier; your exact limits show on the console Limits page. Developer tier needs a payment method.

Source, Nov 15, 2026
No, API needs pay-as-you-go

Free mode gives a monthly budget to try Vibe and Studio but no API access; API needs billing activation.

Source, Nov 15, 2026
Yes

25+ free models, 20 requests/min and 50/day (1,000/day after $10 in credits).

Source, Nov 15, 2026
Starting price
$0.075 per 1M input tokens

GPT-OSS 20B; $0.30 per 1M output tokens. GPT-OSS 120B is $0.15 in, $0.60 out.

Source, Nov 15, 2026
$0.70 per 1M input tokens

Mistral Large 4; $2.10 per 1M output tokens.

Source, Nov 15, 2026
5.5% fee on credits

Standard pay-as-you-go plan; model usage billed at provider rates.

Source, Nov 15, 2026
Rate limit30 requests/min

Developer plan base limits, openai/gpt-oss-120b; 1K requests/day, 8K tokens/min.

Source, Nov 15, 2026
Not listed20 requests/min

Free (:free) models; 50/day under $10 purchased. No plan limit on paid models.

Source, Nov 15, 2026
Size limit131K token context

GPT-OSS 120B (Whisper speech-to-text: 100 MB max file).

Source, Nov 15, 2026
1M token context

Mistral Large 4.

Source, Nov 15, 2026
Not listed
Official SDKsPythonJavaScript / TypeScriptSource, Nov 15, 2026PythonJavaScript / TypeScriptPythonJavaScript / TypeScript
Open sourceNot listedPartly

Open-weight models such as Small 4 and Ministral 3 under Apache 2.0; Large 4 under Mistral's own open-weight license.

Source, Nov 15, 2026
Not listed
Trains on your dataNot listedOpt-out

Studio/API inputs and outputs may be used for training; customers can opt out anytime.

Source, Nov 15, 2026
No

OpenRouter doesn't train on inputs/outputs; some providers may. Models that don't train are labeled.

Source, Nov 15, 2026
EU hostingNo

Retained customer data is stored in GCP buckets in the United States.

Source, Nov 15, 2026
Yes

Data hosted in the EU by default; US endpoint optional.

Source, Nov 15, 2026
Yes

EU in-region routing via eu.openrouter.ai on Business and Enterprise plans.

Source, Nov 15, 2026
ComplianceSOC 2GDPRHIPAA

Trust Center lists SOC 2 Type II, GDPR and HIPAA, with SOC 2 Type II reports for 2024 and 2025.

Source, Nov 10, 2026
SOC 2ISO 27001Source, Nov 15, 2026SOC 2GDPRSource, Nov 15, 2026
Source, Nov 15, 2026
Go
Source, Nov 15, 2026