OpenAI compatible API · Attested · Public status

Meta: Llama 3.3 70B Instruct vs OpenAI: gpt-oss-20b

Compare Meta: Llama 3.3 70B Instruct and OpenAI: gpt-oss-20b by providers, context, price, and TrustedRouter route support.

Verify gateway

Onebase URL to migrate

100sof models and routes

0prompt or output logs. Always.

Compare routesProviders, price, context, and policy posture in one view.

Use auto when uptime mattersKeep a primary model and let fallback handle provider failures.

Same API shapeUse the OpenAI client and set the model you want.

Practical read

OpenAI: gpt-oss-20b has the lower cheapest prompt+completion route on TrustedRouter. Meta: Llama 3.3 70B Instruct has more provider fallback routes, while Meta: Llama 3.3 70B Instruct has the larger context window. Probe-backed speed data is shown when enough recent samples exist.

$0.56175/1MMeta: Llama 3.3 70B Instruct cheapest route

$0.1785/1MOpenAI: gpt-oss-20b cheapest route

not enough dataMeta: Llama 3.3 70B Instruct measured p50 TTFT

not enough dataOpenAI: gpt-oss-20b measured p50 TTFT

Meta: Llama 3.3 70B InstructOpenAI: gpt-oss-20b

Model idmeta-llama/llama-3.3-70b-instructopenai/gpt-oss-20b

PublisherCerebrasOpenAI

AI IQ — IQ 100#69

Context131,072 tokens131,072 tokens

Provider routes1414

Cheapest route$0.56175/1M$0.1785/1M

Privacy posturehas provider E2EE routehas ZDR route

Modes chat chat

Meta: Llama 3.3 70B Instruct routes

Overview14 endpoints
PricingPrompt and completion rates
BenchmarksTrustedRouter and external sources
ProvidersAll serving providers

OpenAI: gpt-oss-20b routes

Overview14 endpoints
PricingPrompt and completion rates
ProvidersAll serving providers

Production choice

Pick a default model. Keep fallback enabled.

TrustedRouter is useful when you know the model you want, but still need provider rollover, budget limits, usage records, and a prompt path you can verify.

OpenAI clientPython

client = OpenAI(
    base_url="https://gchircrcif.eu-west-3.awsapprunner.com/v1",
    api_key="sk-tr-v1-..."
)

response = client.chat.completions.create(
    model="meta-llama/llama-3.3-70b-instruct",
    messages=messages,
)