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 Coder 480B A35B Instruct
Kimi K3 vs Qwen3 Coder 480B A35B Instruct
Output tokens cost $15.00 per million on Kimi K3 and $3.28 per million on Qwen3 Coder 480B A35B Instruct. Input tokens cost $3.00 per million on Kimi K3 and $0.82 per million on Qwen3 Coder 480B A35B Instruct. Cached input tokens are billed at $0.30 per million on Kimi K3 and $0.82 per million on Qwen3 Coder 480B A35B Instruct. Context lengths are 1,048,576 tokens on Kimi K3 and 262,000 tokens on Qwen3 Coder 480B A35B Instruct. Time to first token (TTFT) measured on AIHubMix is 0.9s on Kimi K3 and 0.9s on Qwen3 Coder 480B A35B Instruct. Measured output throughput is 73.1 tok/s on Kimi K3 and 1655.0 tok/s on Qwen3 Coder 480B A35B Instruct. On the LMArena coding leaderboard Kimi K3 scores 1531 and Qwen3 Coder 480B A35B Instruct scores 1457.
Kimi K3Qwen3 Coder 480B A35B Instruct+ Add model · 2/4
Qwen3 Coder 480B A35B Instruct
Qwen · text → text
The code generation model based on Qwen3 has powerful Coding Agent capabilities, achieving state-of-the-art performance compared to open-source models.The model adopts tiered pricing.
Input$0.82 /M
Output$3.28 /M
Pricing & Specifications
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Kimi K3
Qwen3 Coder 480B A35B Instruct
Input /M
$3.00
$0.82
Output /M
$15.00
$3.28
Cache read /M
$0.30
$0.82
Context length
1,048,576
262,000
Max output
1,048,576
262,000
Time to First Token
0.9 s
0.9 s
Throughput
73.1 tok/s
1655.0 tok/s
Modalities
textimagevideo
text
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-coder-480b-a35b-instruct
Tokens / day
-
Requests / day
-
Performance Past 3 Days
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
kimi-k3qwen3-coder-480b-a35b-instruct
Throughput (tok/s)
-
TTFT (s)
-
Uptime (%)
-
LMArena Benchmarks
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Text
kimi-k3qwen3-coder-480b-a35b-instruct
1340140014601520
Overall
13881486
Coding
14571531
Math
13771377
Hard prompts
14141505
Instruction following
13851479
Multi-turn
13991499
Creative writing
13661470
Longer query
14091502
Chinese
14071527
English
14001491
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 Coder 480B A35B Instruct
$98.40 /mo
Kimi K3
$405 /mo
Monthly = daily × 30. Discounted rates applied where a promotion is active.
FAQ
Which is cheaper: Kimi K3, Qwen3 Coder 480B A35B Instruct?
Qwen3 Coder 480B A35B Instruct: $3.28/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 Coder 480B A35B Instruct: 1457 (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.
How large is each context window?
Kimi K3 accepts 1,048,576 and Qwen3 Coder 480B A35B Instruct accepts 262,000 input tokens. Maximum output per request is 1,048,576 tokens on Kimi K3 and 262,000 tokens on Qwen3 Coder 480B A35B Instruct.
Which one generates tokens faster?
Qwen3 Coder 480B A35B Instruct at 1655.0 tok/s and Kimi K3 at 73.1 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 Coder 480B A35B Instruct accepts text input and supports tool calling, function calling and structured outputs.
Can I call Kimi K3 and Qwen3 Coder 480B A35B Instruct 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