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 VL Flash 2026 01-22
Kimi K3 vs Qwen3 VL Flash 2026 01-22
Output tokens cost $15.00 per million on Kimi K3 and $0.21 per million on Qwen3 VL Flash 2026 01-22. Input tokens cost $3.00 per million on Kimi K3 and $0.02 per million on Qwen3 VL Flash 2026 01-22. Cached input tokens are billed at $0.30 per million on Kimi K3 and $0.02 per million on Qwen3 VL Flash 2026 01-22. Context lengths are 1,048,576 tokens on Kimi K3 and 254,000 tokens on Qwen3 VL Flash 2026 01-22. On the LMArena coding leaderboard Kimi K3 scores 1531 and Qwen3 VL Flash 2026 01-22 scores 1466.
Kimi K3Qwen3 VL Flash 2026 01-22+ Add model · 2/4
Qwen3 VL Flash 2026 01-22
Qwen · text, image, video → text
The Qwen3 series of compact visual-understanding models achieves an effective fusion of thinking mode and non-thinking mode, outperforming the open-source Qwen3-VL-30B-A3B with faster response speeds. It comprehensively upgrades image and video understanding, supporting ultra-long contexts such as long videos and long documents, spatial awareness, and universal object recognition; it also possesses visual 2D/3D localization capabilities and is capable of handling complex real-world tasks.
Input$0.02 /M
Output$0.21 /M
Pricing & Specifications
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Kimi K3
Qwen3 VL Flash 2026 01-22
Input /M
$3.00
$0.02
Output /M
$15.00
$0.21
Cache read /M
$0.30
$0.02
Context length
1,048,576
254,000
Max output
1,048,576
32,000
Time to First Token
0.9 s
-
Throughput
73.1 tok/s
-
Modalities
textimagevideo
textimagevideo
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-vl-flash-2026-01-22
Tokens / day
-
Requests / day
-
Performance Past 3 Days
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
kimi-k3qwen3-vl-flash-2026-01-22
Throughput (tok/s)
-
TTFT (s)
-
Uptime (%)
-
LMArena Benchmarks
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Text
kimi-k3qwen3-vl-flash-2026-01-22
1340140014601520
Overall
14151486
Coding
14661531
Math
14111411
Hard prompts
14401505
Instruction following
14141479
Multi-turn
14261499
Creative writing
13621470
Longer query
14251502
Chinese
14551527
English
14271491
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 VL Flash 2026 01-22
$4.33 /mo
Kimi K3
$405 /mo
Monthly = daily × 30. Discounted rates applied where a promotion is active.
FAQ
Which is cheaper: Kimi K3, Qwen3 VL Flash 2026 01-22?
Qwen3 VL Flash 2026 01-22: $0.21/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 VL Flash 2026 01-22: 1466 (LMArena coding leaderboard).
How large is each context window?
Kimi K3 accepts 1,048,576 and Qwen3 VL Flash 2026 01-22 accepts 254,000 input tokens. Maximum output per request is 1,048,576 tokens on Kimi K3 and 32,000 tokens on Qwen3 VL Flash 2026 01-22.
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 VL Flash 2026 01-22 accepts text, image and video input and supports tool calling, function calling and structured outputs.
Can I call Kimi K3 and Qwen3 VL Flash 2026 01-22 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