Kimi K3 vs GLM-5.3
Kimi K3 comes out ahead on 9 of 12 shared measures, and the two rank 14 and 15 of 51 overall. GLM-5.3 is 2.8× cheaper per million tokens (blended) — A$5.78 vs A$14.74 per 1,000 replies of customer support replys.
Where each one wins.
GLM-5.3 ranks 14th of 51; Kimi K3 ranks 15th.
Kimi K3 ranks 5th of 42; GLM-5.3 ranks 11th.
GLM-5.3 ranks 8th of 41; Kimi K3 ranks 9th.
Kimi K3 ranks 11th of 51; GLM-5.3 ranks 15th.
Kimi K3 ranks 1st of 51; GLM-5.3 ranks 31st.
GLM-5.3 ranks 10th of 50; Kimi K3 ranks 27th.
GLM-5.3 is 2.8× cheaper per million tokens (blended) — A$5.78 vs A$14.74 per 1,000 replies of customer support replys.
GLM-5.3 writes 1.7× faster (63 tok/s vs 38 tok/s).
Both can be used from Australia through their makers’ own APIs (Moonshot AI and Z.ai), which may process requests outside Australia. 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. GLM-5.3’s weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region.
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.
| BENCHMARK | KIMI K3 | GLM-5.3 | LEADS |
|---|---|---|---|
| AA Intelligence Index | 43.6 | 44.8 | GLM-5.3 |
| AA Coding Index | 76.2 | 74.8 | Kimi K3 |
| GPQA Diamond | 93.5% | 91.7% | Kimi K3 |
| Humanity’s Last Exam | 46.9% | 42.3% | Kimi K3 |
| AA-LCR | 88.7% | 79.7% | Kimi K3 |
| Terminal-Bench 2.1 | 85.0% | 83.9% | Kimi K3 |
| τ²-Bench (banking) | 46.0% | 50.3% | GLM-5.3 |
| SciCode | 59.5% | 59.0% | Kimi K3 |
| LMArena Text | 1485 | 1483 | Level |
| LMArena WebDev | 1658 | 1620 | Kimi K3 |
| LMArena Agent | #8 | #18 | Kimi K3 |
| Blended price (AUD) | A$8.42 | A$3.02 | GLM-5.3 |
| Output speed | 38 tok/s | 63 tok/s | GLM-5.3 |
| Answer starts after | 56 s | 35 s | GLM-5.3 |
| Context window | 1,048,576 tokens | 1,000,000 tokens | Kimi K3 |
| USE FROM AUSTRALIA | |||
| Maker’s API | YES | YES | |
| Processed in Australia | SELF-HOST ONLY | SELF-HOST ONLY | |
| AUSTRALIAN CLOUD REGIONS | |||
| AWS Bedrock · Sydney | GLOBAL | NOT OFFERED | |
| AWS Bedrock · Melbourne | GLOBAL | NOT OFFERED | |
| Azure · Australia East | NOT OFFERED | NOT OFFERED | |
| Google Vertex AI · Sydney | NOT OFFERED | NOT OFFERED | |
The same jobs, in Australian dollars.
List prices at the snapshot’s RBA rate, excluding GST and reasoning tokens.
| JOB | KIMI K3 | GLM-5.3 | MEDIAN MODEL |
|---|---|---|---|
| Customer support replyPER 1,000 REPLIES | A$14.74 | A$5.78 | ≈ A$6.58 |
| Summarise a 30-page documentPER 100 DOCUMENTS | A$10.11 | A$4.43 | ≈ A$4.74 |
| Agentic coding taskPER 10 TASKS | A$8.00 | A$3.44 | ≈ A$3.72 |
Each model in full.
Kimi K3 or GLM-5.3?
Is Kimi K3 better than GLM-5.3?
On our overall leaderboard GLM-5.3 ranks 14th and Kimi K3 15th of 51. Which is better depends on the job: the table above compares each benchmark, price and Australian availability.
Which is cheaper, Kimi K3 or GLM-5.3?
GLM-5.3, at A$3.02 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.
Can I use Kimi K3 and GLM-5.3 in Australia?
Both can be used from Australia through their makers’ own APIs (Moonshot AI and Z.ai), which may process requests outside Australia. 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. GLM-5.3’s weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region.
Ratings: LMArena leaderboard dataset (CC BY 4.0), rescaled for the lens scores · leaderboards to 22 Sep 2026
Evaluations, prices and speed: Artificial Analysis (artificialanalysis.ai) · fetched 23 Sep 2026
Exchange rate: Reserve Bank of Australia, table F11.1 · A$1 = US$0.7123 on 22 Sep 2026 · prices exclude GST
Snapshot 23 Sep 2026 · updated weekly · How the rankings work →