OpenAI compatible API · Attested · Public status

MoonshotAI: Kimi K2.6 vs Qwen: Qwen3 235B A22B Thinking 2507

Compare MoonshotAI: Kimi K2.6 and Qwen: Qwen3 235B A22B Thinking 2507 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

Qwen: Qwen3 235B A22B Thinking 2507 has the lower cheapest prompt+completion route on TrustedRouter. MoonshotAI: Kimi K2.6 has more provider fallback routes, while MoonshotAI: Kimi K2.6 has the larger context window. Current TrustedRouter probes show MoonshotAI: Kimi K2.6 with the lower p50 TTFT.

$4.158/1MMoonshotAI: Kimi K2.6 cheapest route

$1.56933/1MQwen: Qwen3 235B A22B Thinking 2507 cheapest route

1555 msMoonshotAI: Kimi K2.6 measured p50 TTFT

2165 msQwen: Qwen3 235B A22B Thinking 2507 measured p50 TTFT

MoonshotAI: Kimi K2.6Qwen: Qwen3 235B A22B Thinking 2507

Model idmoonshotai/kimi-k2.6qwen/qwen3-235b-a22b-thinking-2507

PublisherKimiDeepInfra

AI IQ IQ 119#19 —

Context262,144 tokens262,144 tokens

Provider routes328

Cheapest route$4.158/1M$1.56933/1M

Privacy posturehas provider E2EE routehas confidential-compute route

Modes chat chat

MoonshotAI: Kimi K2.6 routes

Overview32 endpoints
PricingPrompt and completion rates
BenchmarksTrustedRouter and external sources
ProvidersAll serving providers

Qwen: Qwen3 235B A22B Thinking 2507 routes

Overview8 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://api.trustedrouter.com/v1",
    api_key="sk-tr-v1-..."
)

response = client.chat.completions.create(
    model="moonshotai/kimi-k2.6",
    messages=messages,
)