// head_to_head
Extract vs GLM 5.3 Flash
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.
Extract is the cheaper of the two; neither can be ranked on quality here.
Extract 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.
scaledown
Extract
- Blended / 1M
- $0.030
- Context
- 1M
- Released
- Oct 13, 2026
- Overall score
- Not evaluated
z-ai
GLM 5.3 Flash
- Blended / 1M
- $0.245
- Context
- 1.0M
- Released
- Aug 26, 2026
- Overall score
- 71.6
Specs and pricing
| Metric | Extract | GLM 5.3 Flash |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 71.6 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.0161 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.030win | $0.245 |
| Input price / 1M | $0.040 | $0.026win |
| Output price / 1M | Freewin | $0.900 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | — | $0.011 |
| Context window | 1M | 1.0M |
| Max output tokens | 900K | 944K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Extract on top, GLM 5.3 Flash below, both out of 100.
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.
| Workload | Extract | GLM 5.3 Flash |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $9.60/mo | $77.16/mo |
| RAG assistant 8K in / 600 out × 100K requests | $32.00/mo | $68.76/mo |
| Coding agent 40K in / 4K out × 20K requests | $32.00/mo | $84.31/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $40.00/mo | $92.83/mo |
| Bulk classification 500 in / 20 out × 5M requests | $100.00/mo | $147.60/mo |
Which should you pick?
You are cost-constrained
Extract
Cheaper on blended list price at $0.030 per million tokens.
Extract vs GLM 5.3 Flash FAQ
Which is better, Extract or GLM 5.3 Flash?
Extract is the cheaper of the two; neither can be ranked on quality here. Extract 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 Extract cheaper than GLM 5.3 Flash?
Extract is cheaper. On a 3:1 input:output blend, Extract lists at $0.030 per million tokens and GLM 5.3 Flash at $0.245 — Extract is 8.2× cheaper. Input and output are priced separately — Extract charges $0.040 in and Free out, GLM 5.3 Flash charges $0.026 and $0.900 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Does Extract or GLM 5.3 Flash have a bigger context window?
They are effectively the same — 1M for Extract and 1.0M for GLM 5.3 Flash.
Do Extract and GLM 5.3 Flash support prompt caching?
GLM 5.3 Flash publishes a cached-input rate of $0.011 per million tokens against a full input rate of $0.026. The catalogue lists no separate cached rate for Extract, which means the provider does not price it separately here — not that caching is unavailable.
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.
- Scores — LiveBench 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.