Skip to content
Model Pareto

GLM-5.3 Flash

Z.ai (Zhipu)Open weightsz.ai ↗
Compare with nearest rivalsCompare with…
Input
$0.15/M tok
Output
$0.50/M tok
Blended
$0.237/M tok · 3:1
Speed
46tok/s
Context
1Mtokens

Open weights (320B-A18B, MIT), natively multimodal.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.

Hardware
4× B200 180GB
FP8 weights
Throughput
~16,470 tok/s
many requests batched
Price
$0.16–1.45 /M tok
busy → light use
On your own machine
Mac Studio M5 Ultra 512GB (int4, ~93 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (320B params, 18B active per token) + 40% KV-cache headroom ≈ 448 GB
  • 4× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~16,470 output tok/s aggregate at batch 471 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~236,250 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

SWE-bench VerifiedEstimated
~#3 of 14 (estimated)~84.6%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3GLM-5.3 FlashEstimated84.6%
  4. #3Gemini 3.1 Pro80.6%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#6 of 19 (estimated)~66.6%
  1. #1Claude Opus 5.589.9%
  2. #2Claude Sonnet 5.581.3%
  3. #5Intern-S2-397B68.5%
  4. ~#6GLM-5.3 FlashEstimated66.6%
  5. #6Hy4 preview65.7%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#3 of 20 (estimated)~91.5%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. ~#3GLM-5.3 FlashEstimated91.5%
  4. #3Qwen3.8 27B90.3%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~#10 of 15 (estimated)~49.0
  1. #1Claude Opus 5.566.0
  2. #2Claude Fable 5.162.2
  3. #9Kimi K351.9
  4. ~#10GLM-5.3 FlashEstimated49.0
  5. #10Grok 4.647.0
No published score — estimated from related benchmarks

Writing & chat

IFBenchEstimated
~#17 of 126 (estimated)~75.8%
  1. #1MiniMax-M382.9%
  2. #2Qwen3.8 2.4T A95B82.8%
  3. #15GPT-5.575.9%
  4. ~#17GLM-5.3 FlashEstimated75.8%
  5. #17Qwen3.5 122B A10B75.7%
No published score — estimated from related benchmarks
MMMLUEstimated
~#2 of 5 (estimated)~90.7%
  1. #1Gemini 3.1 Pro92.6%
  2. ~#2GLM-5.3 FlashEstimated90.7%
  3. #2Gemma 4 31B88.4%
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Verified
#11 of 531463 Elo
  1. #1Claude Opus 5.51501
  2. #2Claude Fable 5.1Few votes1496
  3. #10GLM-5.31466
  4. #11GLM-5.3 Flash1463
  5. #12DeepSeek V4.1 Flash1462
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#14 of 531475 Elo
  1. #1Claude Opus 5.51509
  2. #2Claude Fable 5.11501
  3. #13DeepSeek V4.1 Flash1477
  4. #14GLM-5.3 Flash1475
  5. #15GPT-5.6 Terra1466
lmarena.ai · 2026-09-24
EQ-Bench Creative Writing v3 (Elo)Estimated
~#9 of 26 (estimated)~1921 Elo
  1. #1GPT-6 Astra2173
  2. #2Claude Fable 5.12162
  3. #8Grok 4.72007
  4. ~#9GLM-5.3 FlashEstimated1921
  5. #9Muse Spark 1.31906
No published score — estimated from related benchmarks

Vision

MMMUEstimated
~#2 of 3 (estimated)~71.3%
  1. #1Claude Haiku 4.573.2%
  2. ~#2GLM-5.3 FlashEstimated71.3%
  3. #2Llama 4 Maverick69.4%
No published score — estimated from related benchmarks
MMMU-ProEstimated
~#7 of 77 (estimated)~82.5%
  1. #1Claude Opus 5.587.7%
  2. #2GPT-6 Astra86.9%
  3. #6Qwen3.8 Max (0902)82.8%
  4. ~#7GLM-5.3 FlashEstimated82.5%
  5. #7Gemini 3.1 Pro82.4%
No published score — estimated from related benchmarks
CharXiv ReasoningEstimated
~#5 of 9 (estimated)~81.3%
  1. #1Gemini 3.8 Flash86.2%
  2. #2Kimi K384.8%
  3. #4Qwen3.8-Omni-Flash83.5%
  4. ~#5GLM-5.3 FlashEstimated81.3%
  5. #5Muse Glimmer78.8%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#4 of 8 (estimated)~87.5%
  1. #1MiniMax-M391.6%
  2. #2Kimi K391.1%
  3. #2Qwen3.8 27B91.1%
  4. ~#4GLM-5.3 FlashEstimated87.5%
  5. #4Gemini 3.1 Pro85.3%
No published score — estimated from related benchmarks

Hard reasoning

AIME (latest)Estimated
~#2 of 25 (estimated)~96.7%
  1. #1Inkling97.1%
  2. ~#2GLM-5.3 FlashEstimated96.7%
  3. #2Muse Glimmer94.7%
No published score — estimated from related benchmarks

Agents

τ²-bench (Telecom)Estimated
~#2 of 116 (estimated)~97.4%
  1. #1Step 3.7 Flash98.5%
  2. ~#2GLM-5.3 FlashEstimated97.4%
  3. #2Gemini 3.1 Pro95.6%
No published score — estimated from related benchmarks
OSWorld-VerifiedEstimated
~#3 of 12 (estimated)~82.9%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. ~#3GLM-5.3 FlashEstimated82.9%
  4. #3Nex-N2.5-Pro82.2%
No published score — estimated from related benchmarks
OSWorld 2.0Lab-reported
#1 of 459.1%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
z.ai · 2026-08-26
BrowseCompEstimated
~#5 of 18 (estimated)~89.8%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #4Claude Opus 590.8%
  4. ~#5GLM-5.3 FlashEstimated89.8%
  5. #5Nex-N2.5-Pro89.7%
No published score — estimated from related benchmarks