Kimi K3
Moonshot AI · text, image, video → text
Kimi K3 is Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window and industry-leading intelligence.
Input$3.00 /M
Output$15.00 /M
← ModelsKimi K3
Qwen3 Max Preview
Kimi K3 vs Qwen3 Max Preview
Output tokens cost $15.00 per million on Kimi K3 and $3.38 per million on Qwen3 Max Preview. Input tokens cost $3.00 per million on Kimi K3 and $0.85 per million on Qwen3 Max Preview. Cached input tokens are billed at $0.30 per million on Kimi K3 and $0.85 per million on Qwen3 Max Preview. Time to first token (TTFT) measured on AIHubMix is 0.9s on Kimi K3 and 1.2s on Qwen3 Max Preview. Measured output throughput is 73.1 tok/s on Kimi K3 and 7.4 tok/s on Qwen3 Max Preview. On the LMArena coding leaderboard Kimi K3 scores 1531 and Qwen3 Max Preview scores 1482.
Kimi K3Qwen3 Max Preview+ Add model · 2/4
Qwen3 Max Preview
Qwen · text, image → text
Qwen3-Max-Preview is the latest preview model in the Qwen3 series. This version is functionally equivalent to Qwen3-Max-Thinking — simply set extra_body={"enable_thinking": True} to enable the thinking mode. Compared to the Qwen2.5 series, it delivers significant improvements in overall general capabilities, including English–Chinese text understanding, complex instruction following, open-ended reasoning, multilingual processing, and tool-use proficiency. The model also exhibits fewer hallucinations and stronger overall reliability.
Input$0.85 /M
Output$3.38 /M
Pricing & Specifications
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Kimi K3
Qwen3 Max Preview
Input /M
$3.00
$0.85
Output /M
$15.00
$3.38
Cache read /M
$0.30
$0.85
Context length
1,048,576
0
Max output
1,048,576
0
Time to First Token
0.9 s
1.2 s
Throughput
73.1 tok/s
7.4 tok/s
Modalities
textimagevideo
textimage
Supported Parameters
thinkingfunction callingstructured outputs
toolsfunction callingstructured outputs
API Formats
Released
-
-
Promotional prices show the discounted rate; see each model page for promotion windows.
Activity Past 30 Days
Daily traffic served through AIHubMix — how demand for each model is trending.
kimi-k3qwen3-max-preview
Tokens / day
-
Requests / day
-
Performance Past 3 Days
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
kimi-k3qwen3-max-preview
Throughput (tok/s)
-
TTFT (s)
-
Uptime (%)
-
LMArena Benchmarks
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Text
kimi-k3qwen3-max-preview
1360142014801540
Overall
14351486
Coding
14821531
Math
14381438
Hard prompts
14591505
Instruction following
14271479
Multi-turn
14451499
Creative writing
13941470
Longer query
14441502
Chinese
14891527
English
14391491
Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.
Cost calculator
Estimate your monthly bill for the same workload on each model.
Qwen3 Max Preview
$102 /mo
Kimi K3
$405 /mo
Monthly = daily × 30. Discounted rates applied where a promotion is active.
FAQ
Which is cheaper: Kimi K3, Qwen3 Max Preview?
Qwen3 Max Preview: $3.38/M output tokens; Kimi K3: $15.00/M. Use the cost calculator above to estimate your own workload.
How do their coding arena scores compare?
Kimi K3: 1531; Qwen3 Max Preview: 1482 (LMArena coding leaderboard).
Which responds faster?
Kimi K3: 0.9s time to first token measured on AIHubMix; see the live performance charts above for how each model behaves across the day.
Which one generates tokens faster?
Kimi K3 at 73.1 tok/s and Qwen3 Max Preview at 7.4 tok/s, measured as output throughput on AIHubMix — a separate metric from time to first token.
What inputs and capabilities does each model support?
Kimi K3 accepts text, image and video input and supports thinking, function calling and structured outputs; Qwen3 Max Preview accepts text and image input and supports tool calling, function calling and structured outputs.
Can I call Kimi K3 and Qwen3 Max Preview with the same API key?
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.
Popular comparisons
Related model match-ups readers also look at.
AIHubMix