Output tokens cost $2.19 per million on Kimi K2 Thinking and $1.01 per million on Qwen3.8 Max Preview. Input tokens cost $0.55 per million on Kimi K2 Thinking and $0.34 per million on Qwen3.8 Max Preview. Cached input tokens are billed at $0.14 per million on Kimi K2 Thinking and $0.03 per million on Qwen3.8 Max Preview. Context lengths are 262,144 tokens on Kimi K2 Thinking and 983,616 tokens on Qwen3.8 Max Preview. Time to first token (TTFT) measured on AIHubMix is 0.4s on Kimi K2 Thinking and 2.4s on Qwen3.8 Max Preview. Measured output throughput is 208.3 tok/s on Kimi K2 Thinking and 48.1 tok/s on Qwen3.8 Max Preview. On the LMArena coding leaderboard Kimi K2 Thinking scores 1467 and Qwen3.8 Max Preview scores 1520.
Kimi K2 Thinking
Qwen3.8 Max PreviewKimi K2 Thinking is Moonshot AI's most advanced open-source inference model to date, extending the K2 series into intelligent agent and long-context inference domains. The model is built on the trillion-parameter mixture of experts (MoE) architecture introduced by Kimi K2, activating 32 billion parameters per forward pass and supporting a context window of 256,000 tokens.
Qwen 3.8 Max Preview(Qwen3.8-Max-Preview) is the latest-generation foundation model in the Qwen family, packing 2.4T parameters and still evolving. Compared with the previous flagship Qwen 3.7 Max, it delivers major gains in core capabilities like Coding and Cowork (professional productivity), with world-leading performance on complex, long-horizon tasks such as full-stack development, data analysis, and Office workflows. Launch offer: Credits are consumed at just 20% of the standard rate, effectively 5× your usage. Limited time only.
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Promotional prices show the discounted rate; see each model page for promotion windows.
Daily traffic served through AIHubMix — how demand for each model is trending.
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.
Estimate your monthly bill for the same workload on each model.
Monthly = daily × 30. Discounted rates applied where a promotion is active.
Qwen3.8 Max Preview: $1.01/M output tokens; Kimi K2 Thinking: $2.19/M. Use the cost calculator above to estimate your own workload.
Qwen3.8 Max Preview: 1520; Kimi K2 Thinking: 1467 (LMArena coding leaderboard).
Kimi K2 Thinking: 0.4s time to first token measured on AIHubMix; see the live performance charts above for how each model behaves across the day.
Kimi K2 Thinking accepts 262,144 and Qwen3.8 Max Preview accepts 983,616 input tokens. Maximum output per request is 262,144 tokens on Kimi K2 Thinking and 131,072 tokens on Qwen3.8 Max Preview.
Kimi K2 Thinking at 208.3 tok/s and Qwen3.8 Max Preview at 48.1 tok/s, measured as output throughput on AIHubMix — a separate metric from time to first token.
Kimi K2 Thinking accepts text input and supports thinking, tool calling, function calling and structured outputs; Qwen3.8 Max Preview accepts text and image input and supports tool calling, function calling, structured outputs, web search, long context and thinking.
Yes. AIHubMix serves every model on this page behind one OpenAI-compatible endpoint, so switching between them is a one-line change to the model field — no second account, key or SDK.
Related model match-ups readers also look at.
AIHubMix