Kimi K3 vs Qwen3.8 Max

At a glance

• Capability index

  • Kimi K3157.4
  • Qwen3.8 Max156.4

  Kimi K3 is 1.0 points higher.

• Price per 1M tokens

  • Kimi K3$6.00
  • Qwen3.8 Max$3.00

  Qwen3.8 Max is 2 times cheaper.

• Context window

  • Kimi K31M tokens
  • Qwen3.8 Max1M tokens

  They take in about the same.

• Released

  • Kimi K3Jul 16, 2026
  • Qwen3.8 MaxAug 3, 2026

  Qwen3.8 Max is 18 days newer.

Where each scores higher

On the 22 tests both have taken, each on its own scale. A gap of a point or two can sit within a test’s margin of error.

Everything side by side

Comparison of Kimi K3 vs Qwen3.8 Max

Measure	Kimi K3Moonshot AI	Qwen3.8 MaxAlibaba
Model
Maker	Moonshot AI	Alibaba
Sold here by	Moonshot AI	Alibaba
API model ID	kimi-k3	qwen3.8-max
Released	Jul 16, 2026	Aug 3, 2026
Knowledge cutoff	Not reported	Not reported
Weights	Open: downloadable	Closed: hosted access only
Developer’s country	China	China
Hosts selling it	84including Moonshot AI directly	36including Alibaba directly
Training compute	2.0 × 10²⁵ FLOPEpoch AI estimate	Not published
Test results
Capability index	157.4 (best of these models)#15 of 274	156.4#22 of 274
LiveBench overall	79.2% (best of these models)#15 of 66	78.5%#17 of 66
Arena text	1488 (best of these models)#16 of 413	1482#23 of 413
LiveBench coding	81.5% (best of these models)#15 of 66	72.9%#52 of 66
Arena WebDev	1658#13 of 138	1671 (best of these models)#10 of 138
Arena coding	1541 (best of these models)#9 of 408	1524#28 of 408
LiveBench mathematics	84.4%#53 of 66	91.3% (best of these models)#27 of 66
Mock AIME 2024–2025	97.2%#31 of 297	99.4% (best of these models)#14 of 297
FrontierMath Tiers 1–3	72.2%#22 of 114	74.7% (best of these models)#18 of 114
FrontierMath Tier 4	39.0%#29 of 70	46.3% (best of these models)#25 of 70
Arena math	1501 (best of these models)#16 of 396	1498#18 of 396
SimpleQA Verified	50.6% (best of these models)#24 of 86	45.8%#37 of 86
LiveBench agentic coding	62.2%#10 of 66	64.7% (best of these models)#7 of 66
DeepSWE	68.5% (best of these models)#19 of 69	57.5%#36 of 69
LiveBench language	85.5% (best of these models)#11 of 66	79.7%#34 of 66
LiveBench instructions	71.4%#23 of 66	74.1% (best of these models)#13 of 66
Arena creative writing	1460#26 of 411	1468 (best of these models)#20 of 411
Arena instruction following	1488 (best of these models)#14 of 413	1474#31 of 413
LiveBench data analysis	78.7% (best of these models)#21 of 66	78.4%#24 of 66
LiveBench reasoning	90.7% (best of these models)#9 of 66	88.2%#23 of 66
GPQA Diamond	93.1% (best of these models)#20 of 319	92.7%#24 of 319
Chess puzzles	39.0% (best of these models)#31 of 227	29.0%#54 of 227
Arena hard prompts	1517 (best of these models)#9 of 413	1503#27 of 413
Price per million tokens
Input	$3.00	$2.00 (best of these models)
Output	$15.00	$6.00 (best of these models)
Blended, 3 input to 1 output	$6.00	$3.00 (best of these models)
Cached input	$0.30	$0.25 (best of these models)
Writing to the cache	Not reported	$2.50
The maker’s own price	This listing	This listing
Limits
Context window	1M tokens	1M tokens
Max input	1M tokens (best of these models)	991K tokens
Max output	1M tokens (best of these models)	131.1K tokens
Features
Reads	images, text and video	images, PDFs, text and video
Produces	text	text
Reasoning	Yeseffort low, high, max; can be switched off; thinking budget can be set	Yeseffort low, medium, xhigh; can be switched off; thinking budget can be set
Tool calling	Yes	Yes
Structured output	Yes, such as JSON	Yes, such as JSON
File attachments	Yes	Yes
Temperature setting	Fixed by the provider	Supported

Bold marks the better value in each row. Prices are each model’s own list price, or the middle price across hosts when the maker does not sell it, blended as three input tokens for every output token. Test results as published by LiveBench (Apache-2.0), Epoch AI (CC BY 4.0), Datacurve DeepSWE, compiled by Epoch AI (CC BY 4.0), Arena (CC BY 4.0); model details and prices from models.dev (MIT). Confirm pricing with the provider before use.