ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GLM-4 32B 0414 128KvsGLM-4.5-Air

GLM-4 32B 0414 128K and GLM-4.5-Air do not yet share a public benchmark, so quality cannot be compared apples-to-apples. GLM-4 32B 0414 128K is 6.5x cheaper per token. GLM-4 32B 0414 128K has the larger context window (131K tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
0
no overlap yet
Cheaper per token
GLM-4 32B 0414 128K
6.5x cheaper, input + output
Larger context
GLM-4 32B 0414 128K
131K tokens
Provider uptime (30d)
Z.ai
GLM-4 32B 0414 128K
Z.ai · released 2025-04-14
text
Context
131K
Max out
16K
Input /1M
$0.1
Output /1M
$0.1
Cached /1M
Scores
0 · 0 core
text
Context
128K
Max out
98K
Input /1M
$0.2
Output /1M
$1.1
Cached /1M
$0.03
Scores
2 · 2 core
Quality

Benchmark matrix

BenchmarkGLM-4 32B 0414 128KGLM-4.5-AirΔEdge
Reported by both · 0
No public benchmark is reported for both models yet.
Only GLM-4 32B 0414 128K reports · 0
GLM-4 32B 0414 128K reports nothing GLM-4.5-Air does not.
Only GLM-4.5-Air reports · 2
AIME 2025not reported83.3% matharena
LMArena Elonot reported1383.6
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GLM-4 32B 0414 128K minus GLM-4.5-Air in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecGLM-4 32B 0414 128KGLM-4.5-AirEdge
Context window131K tokens128K tokensGLM-4 32B 0414 128K
Max output16K tokens98K tokensGLM-4.5-Air
Input price / 1M$0.1 $0.2 GLM-4 32B 0414 128K
Output price / 1M$0.1 $1.1 GLM-4 32B 0414 128K
Cached input / 1M$0.03
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$0.2$1.3GLM-4 32B 0414 128K
ModalitiestexttextTie
Released2025-04-14
Cited benchmark scores02
Reliability

Z.ai status

All providers
Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GLM-4 32B 0414 128K and GLM-4.5-Air on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.