// head_to_head

Qwen3.8 27B vs GLM 5.2

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

Qwen3.8 27B wins outright — it scores higher and costs less.

Qwen3.8 27B leads by 2.1 points overall while listing 30% cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. Qwen3.8 27B also leads on measured cost per point of capability, at $0.0556 per point.

qwen

Qwen3.8 27B

Blended / 1M
$1.14
Context
1M
Released
Aug 14, 2026
Overall score
75.3
reasoningtool callingimage inputvideo inputprompt caching

z-ai

GLM 5.2

Blended / 1M
$1.48
Context
1.0M
Released
Jun 16, 2026
Overall score
73.2
reasoningtool callingprompt caching

Specs and pricing

MetricQwen3.8 27BGLM 5.2
LiveBench overall

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

75.3win73.2
Cost per point

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

$0.0556win$0.1260
Blended price / 1M

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

$1.14win$1.48
Input price / 1M$0.450win$0.966
Output price / 1M$3.20$3.04
Cached input / 1M

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

$0.050win$0.193
Context window1M1.0M
Max output tokens131K131K

Benchmarks by category

An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Qwen3.8 27B on top, GLM 5.2 below, both out of 100.

Agentic coding
61.4
51.8
Coding
75.7
79.7
Reasoning
80.0
78.6
Mathematics
86.2
89.8
Data analysis
76.6
73.7
Language
74.3
76.2
Instruction following
72.7
62.3

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.

WorkloadQwen3.8 27BGLM 5.2
Support chatbot

1.2K in / 400 out × 200K requests

$335.20/mo$419.08/mo
RAG assistant

8K in / 600 out × 100K requests

$392.00/mo$645.84/mo
Coding agent

40K in / 4K out × 20K requests

$392.00/mo$582.91/mo
Document extraction

20K in / 1.5K out × 50K requests

$670.00/mo$1155.06/mo
Bulk classification

500 in / 20 out × 5M requests

$1245.00/mo$2332.20/mo
Run these two through the cost calculator

Which should you pick?

You are running this at volume

Qwen3.8 27B

Lowest measured cost per point of capability at $0.0556 per point — the gap compounds with every request.

Quality matters more than the bill

Qwen3.8 27B

Highest overall LiveBench score of the two at 75.3.

The workload is coding or agentic work

Qwen3.8 27B

Leads on agentic coding — 61.4 against 51.8.

Qwen3.8 27B vs GLM 5.2 FAQ

Which is better, Qwen3.8 27B or GLM 5.2?

Qwen3.8 27B wins outright — it scores higher and costs less. Qwen3.8 27B leads by 2.1 points overall while listing 30% cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. Qwen3.8 27B also leads on measured cost per point of capability, at $0.0556 per point.

Is Qwen3.8 27B cheaper than GLM 5.2?

Qwen3.8 27B is cheaper. On a 3:1 input:output blend, Qwen3.8 27B lists at $1.14 per million tokens and GLM 5.2 at $1.48 — Qwen3.8 27B is 30% cheaper. Input and output are priced separately — Qwen3.8 27B charges $0.450 in and $3.20 out, GLM 5.2 charges $0.966 and $3.04 — so the model that looks cheaper flips depending on how output-heavy your workload is.

Qwen3.8 27B vs GLM 5.2: which scores higher on benchmarks?

Qwen3.8 27B scores 75.3 and GLM 5.2 scores 73.2 overall on LiveBench, the mean of its seven categories. That is a 2.1-point lead for Qwen3.8 27B. Category scores differ from the overall figure — a model can lead on reasoning and trail on coding, which the per-category table above breaks out.

Which gives better value for money, Qwen3.8 27B or GLM 5.2?

Qwen3.8 27B. Cost per point divides the measured dollars LiveBench spent running the benchmark by the score it earned, so it captures something token pricing misses: a reasoning model can emit many times more tokens than its per-token rate suggests. Qwen3.8 27B works out at $0.0556 per point and GLM 5.2 at $0.1260.

Does Qwen3.8 27B or GLM 5.2 have a bigger context window?

They are effectively the same — 1M for Qwen3.8 27B and 1.0M for GLM 5.2.

Do Qwen3.8 27B and GLM 5.2 support prompt caching?

Both publish a cached-input rate: $0.050 per million for Qwen3.8 27B and $0.193 for GLM 5.2, against full input rates of $0.450 and $0.966. 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.