// head_to_head

Gemini 2.5 Flash vs GLM 5.3

Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.

OpenRouter + LiveBenchAll comparisonsFull leaderboard

Gemini 2.5 Flash is the cheaper of the two; neither can be ranked on quality here.

Gemini 2.5 Flash does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.

google

Gemini 2.5 Flash

Blended / 1M
$0.850
Context
1.0M
Released
Jun 17, 2025
Overall score
Not evaluated
reasoningtool callingfile inputimage inputaudio inputvideo inputprompt caching

z-ai

GLM 5.3

Blended / 1M
$1.29
Context
1.3M
Released
Aug 18, 2026
Overall score
76.1
reasoningtool callingprompt caching

Specs and pricing

MetricGemini 2.5 FlashGLM 5.3
LiveBench overall

Mean of the seven LiveBench category scores, 0–100. Higher is better.

76.1
Cost per point

Measured benchmark spend divided by overall score — dollars per point of capability.

$0.2460
Blended price / 1M

3:1 input:output mix, the usual shape of production traffic.

$0.850win$1.29
Input price / 1M$0.300win$0.840
Output price / 1M$2.50$2.64
Cached input / 1M

Price of an input token served from the prompt cache, where the provider publishes one.

$0.030win$0.156
Context window1.0M1.3M
Max output tokens66K131Kwin

Benchmarks by category

An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Gemini 2.5 Flash on top, GLM 5.3 below, both out of 100.

Agentic coding
60.9
Coding
79.0
Reasoning
85.8
Mathematics
87.9
Data analysis
70.2
Language
79.9
Instruction following
69.3

What each one costs to run

Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.

WorkloadGemini 2.5 FlashGLM 5.3
Support chatbot

1.2K in / 400 out × 200K requests

$252.56/mo$363.55/mo
RAG assistant

8K in / 600 out × 100K requests

$282.00/mo$556.80/mo
Coding agent

40K in / 4K out × 20K requests

$288.80/mo$500.16/mo
Document extraction

20K in / 1.5K out × 50K requests

$474.00/mo$1,004/mo
Bulk classification

500 in / 20 out × 5M requests

$865.00/mo$2,022/mo
Run these two through the cost calculator

Which should you pick?

You are cost-constrained

Gemini 2.5 Flash

Cheaper on blended list price at $0.850 per million tokens.

Gemini 2.5 Flash vs GLM 5.3 FAQ

Which is better, Gemini 2.5 Flash or GLM 5.3?

Gemini 2.5 Flash is the cheaper of the two; neither can be ranked on quality here. Gemini 2.5 Flash does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.

Is Gemini 2.5 Flash cheaper than GLM 5.3?

Gemini 2.5 Flash is cheaper. On a 3:1 input:output blend, Gemini 2.5 Flash lists at $0.850 per million tokens and GLM 5.3 at $1.29 — Gemini 2.5 Flash is 1.5× cheaper. Input and output are priced separately — Gemini 2.5 Flash charges $0.300 in and $2.50 out, GLM 5.3 charges $0.840 and $2.64 — so the model that looks cheaper flips depending on how output-heavy your workload is.

Does Gemini 2.5 Flash or GLM 5.3 have a bigger context window?

They are effectively the same — 1.0M for Gemini 2.5 Flash and 1.3M for GLM 5.3.

Do Gemini 2.5 Flash and GLM 5.3 support prompt caching?

Both publish a cached-input rate: $0.030 per million for Gemini 2.5 Flash and $0.156 for GLM 5.3, against full input rates of $0.300 and $0.840. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.

Related comparisons

How these numbers are produced

  • Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
  • ScoresLiveBench release 2026-06-25, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
  • Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
  • “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.

Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.