Qwen3.8 Max vs Kimi K3

Kimi K3 comes out ahead on 6 of 12 shared measures, and the two rank 15 and 16 of 51 overall. Qwen3.8 Max is 2.0× cheaper per million tokens (blended) — A$8.14 vs A$14.74 per 1,000 replies of customer support replys.

QWEN3.8 MAX #16KIMI K3 #15UPDATED 23 SEP 2026
§ 01 — AT A GLANCE

Where each one wins.

Overall

Kimi K3 ranks 15th of 51; Qwen3.8 Max ranks 16th.

Coding

Qwen3.8 Max ranks 4th of 42; Kimi K3 ranks 5th.

Agents

Qwen3.8 Max ranks 7th of 41; Kimi K3 ranks 9th.

Reasoning

Kimi K3 ranks 11th of 51; Qwen3.8 Max ranks 13th.

Long context

Kimi K3 ranks 1st of 51; Qwen3.8 Max ranks 27th.

Value

Qwen3.8 Max ranks 13th of 50; Kimi K3 ranks 27th.

Price

Qwen3.8 Max is 2.0× cheaper per million tokens (blended) — A$8.14 vs A$14.74 per 1,000 replies of customer support replys.

Speed

Qwen3.8 Max writes 1.1× faster (42 tok/s vs 38 tok/s).

Australia

Both can be used from Australia through their makers’ own APIs (Alibaba and Moonshot AI), which may process requests outside Australia. None of the major cloud platforms offer it in an Australian region yet. Kimi K3 can be called from AWS Bedrock in Sydney and Melbourne, but only through global routing, so requests may be processed outside Australia; its weights are public, so it can also be self-hosted in Australia.

§ 02 — EVERY MEASURE

Side by side.

“Level” means the gap is inside LMArena’s confidence interval, or the scores are identical. Independent test scores are for each model’s headline setting. Both models can be used from Australia through their makers’ APIs; the Australian rows show whether requests can also be processed in Australia.

BENCHMARKS, SPEED AND AVAILABILITY23 SEP 2026
Qwen3.8 Max vs Kimi K3: benchmark results
BENCHMARKQWEN3.8 MAXKIMI K3LEADS
AA Intelligence Index45.443.6Qwen3.8 Max
AA Coding Index76.276.2Level
GPQA Diamond92.8%93.5%Kimi K3
Humanity’s Last Exam43.1%46.9%Kimi K3
AA-LCR80.3%88.7%Kimi K3
Terminal-Bench 2.188.8%85.0%Qwen3.8 Max
τ²-Bench (banking)47.8%46.0%Qwen3.8 Max
SciCode52.1%59.5%Kimi K3
LMArena Text14811485Level
LMArena WebDev16711658Level
LMArena Vision1302
LMArena Agent#17#8Kimi K3
Blended price (AUD)A$4.21A$8.42Qwen3.8 Max
Output speed42 tok/s38 tok/sQwen3.8 Max
Answer starts after50 s56 sQwen3.8 Max
Context window1,000,000 tokens1,048,576 tokensKimi K3
USE FROM AUSTRALIA
Maker’s APIYESYES
Processed in AustraliaNOSELF-HOST ONLY
AUSTRALIAN CLOUD REGIONS
AWS Bedrock · SydneyNOT OFFEREDGLOBAL
AWS Bedrock · MelbourneNOT OFFEREDGLOBAL
Azure · Australia EastNOT OFFEREDNOT OFFERED
Google Vertex AI · SydneyNOT OFFEREDNOT OFFERED
§ 03 — COST

The same jobs, in Australian dollars.

List prices at the snapshot’s RBA rate, excluding GST and reasoning tokens.

EVERYDAY JOBSESTIMATES
Estimated cost of common jobs in Australian dollars
JOBQWEN3.8 MAXKIMI K3MEDIAN MODEL
Customer support replyPER 1,000 REPLIESA$8.14A$14.74A$6.58
Summarise a 30-page documentPER 100 DOCUMENTSA$6.29A$10.11A$4.74
Agentic coding taskPER 10 TASKSA$4.89A$8.00A$3.72
§ 05 — QUESTIONS

Qwen3.8 Max or Kimi K3?

On our overall leaderboard Kimi K3 ranks 15th and Qwen3.8 Max 16th of 51. Which is better depends on the job: the table above compares each benchmark, price and Australian availability.

Qwen3.8 Max, at A$4.21 per million tokens (blended) against A$8.42 for Kimi K3. Reasoning-heavy settings add output tokens, so real bills depend on the effort level you run.

Both can be used from Australia through their makers’ own APIs (Alibaba and Moonshot AI), which may process requests outside Australia. None of the major cloud platforms offer it in an Australian region yet. Kimi K3 can be called from AWS Bedrock in Sydney and Melbourne, but only through global routing, so requests may be processed outside Australia; its weights are public, so it can also be self-hosted in Australia.

PUT THE COMPARISON TO WORK

Need help choosing and using AI for your business?