GLM-5.3 vs MiniMax-M3

At a glance

• Capability index

  • GLM-5.3155.6
  • MiniMax-M3146.9

  GLM-5.3 is 8.7 points higher.

• Price per 1M tokens

  • GLM-5.3$2.15
  • MiniMax-M3$0.525

  MiniMax-M3 is 4.1 times cheaper.

• Context window

  • GLM-5.31M tokens
  • MiniMax-M31M tokens

  They take in about the same.

• Released

  • GLM-5.3Aug 14, 2026
  • MiniMax-M3Jun 1, 2026

  GLM-5.3 is 2 months newer.

Where each scores higher

On the 18 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.

Level on GPQA Diamond.

5 more tests have results for only one of them; the table below lists every result.

Everything side by side

Comparison of GLM-5.3 vs MiniMax-M3

Measure	GLM-5.3Z.ai	MiniMax-M3MiniMax
Model
Maker	Z.ai	MiniMax
Sold here by	Z.AI	MiniMax (minimax.io)
API model ID	glm-5.3	MiniMax-M3
Released	Aug 14, 2026	Jun 1, 2026
Knowledge cutoff	Not reported	Jan 2025
Weights	Open: downloadable	Open: downloadable
Developer’s country	China	China
Hosts selling it	75including Z.ai directly	56including MiniMax directly
Test results
Capability index	155.6 (best of these models)#27 of 274	146.9#68 of 274
LiveBench overall	76.1% (best of these models)#32 of 66	67.3%#61 of 66
Arena text	1479 (best of these models)#27 of 413	1440#96 of 413
LiveBench coding	79.0% (best of these models)#26 of 66	68.2%#64 of 66
Arena WebDev	1623 (best of these models)#20 of 138	1482#58 of 138
Arena coding	1521 (best of these models)#34 of 408	1494#86 of 408
LiveBench mathematics	87.9% (best of these models)#43 of 66	77.0%#65 of 66
Mock AIME 2024–2025	91.1% (best of these models)#65 of 297	71.1%#133 of 297
FrontierMath Tiers 1–3	68.8%#25 of 114	Not tested
FrontierMath Tier 4	29.3%#36 of 70	Not tested
Arena math	1494 (best of these models)#20 of 396	1432#102 of 396
SimpleQA Verified	41.0%#44 of 86	Not tested
LiveBench agentic coding	60.9% (best of these models)#15 of 66	40.7%#60 of 66
DeepSWE	69.0%#16 of 69	Not tested
CursorBench	42.6%#20 of 62	Not tested
LiveBench language	79.9% (best of these models)#31 of 66	76.8%#45 of 66
LiveBench instructions	69.3% (best of these models)#33 of 66	57.5%#61 of 66
Arena creative writing	1454 (best of these models)#34 of 411	1406#99 of 411
Arena instruction following	1476 (best of these models)#25 of 413	1434#88 of 413
LiveBench data analysis	70.2%#51 of 66	76.2% (best of these models)#34 of 66
LiveBench reasoning	85.8% (best of these models)#35 of 66	74.5%#61 of 66
GPQA Diamond	90.9%#34 of 319	90.9%#34 of 319
Chess puzzles	21.0% (best of these models)#80 of 227	14.0%#113 of 227
Arena hard prompts	1503 (best of these models)#25 of 413	1463#90 of 413
Price per million tokens
Input	$1.40	$0.30 (best of these models)
Output	$4.40	$1.20 (best of these models)
Blended, 3 input to 1 output	$2.15	$0.525 (best of these models)
Cached input	$0.26	$0.06 (best of these models)
Writing to the cache	Free	Not reported
Long requests	No separate price reported	$0.60 in, $2.40 outabove 512K tokens
The maker’s own price	This listing	This listing
Limits
Context window	1M tokens	1M tokens
Max input	1M tokens (best of these models)	512K tokens
Max output	131.1K tokens	512K tokens (best of these models)
Features
Reads	text	images, text and video
Produces	text	text
Reasoning	Yeseffort low, high, max; can be switched off; thinking budget can be set	Yeseffort low, medium, high; 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	No	Yes
Temperature setting	Supported	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), Cursor CursorBench, 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.