// head_to_head
Gemma 3 27B vs Muse Spark 1.3
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.
Gemma 3 27B is the cheaper of the two; neither can be ranked on quality here.
Gemma 3 27B 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.
Gemma 3 27B
- Blended / 1M
- $0.172
- Context
- 131K
- Released
- Mar 12, 2025
- Overall score
- Not evaluated
meta
Muse Spark 1.3
- Blended / 1M
- $2.00
- Context
- 1.0M
- Released
- Sep 2, 2026
- Overall score
- 81.6
Specs and pricing
| Metric | Gemma 3 27B | Muse Spark 1.3 |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 81.6 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.1210 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.172win | $2.00 |
| Input price / 1M | $0.080win | $1.25 |
| Output price / 1M | $0.450win | $4.25 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.040win | $0.150 |
| Context window | 131K | 1.0Mwin |
| Max output tokens | 118K | 944Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Gemma 3 27B on top, Muse Spark 1.3 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 | Gemma 3 27B | Muse Spark 1.3 |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $52.32/mo | $560.80/mo |
| RAG assistant 8K in / 600 out × 100K requests | $75.00/mo | $815.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $77.60/mo | $724.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $111.75/mo | $1,514/mo |
| Bulk classification 500 in / 20 out × 5M requests | $225.00/mo | $3,000/mo |
Which should you pick?
You need to fit large documents in one call
Muse Spark 1.3
Wider context window — 1.0M against 131K.
You are cost-constrained
Gemma 3 27B
Cheaper on blended list price at $0.172 per million tokens.
Gemma 3 27B vs Muse Spark 1.3 FAQ
Which is better, Gemma 3 27B or Muse Spark 1.3?
Gemma 3 27B is the cheaper of the two; neither can be ranked on quality here. Gemma 3 27B 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 Gemma 3 27B cheaper than Muse Spark 1.3?
Gemma 3 27B is cheaper. On a 3:1 input:output blend, Gemma 3 27B lists at $0.172 per million tokens and Muse Spark 1.3 at $2.00 — Gemma 3 27B is 12× cheaper. Input and output are priced separately — Gemma 3 27B charges $0.080 in and $0.450 out, Muse Spark 1.3 charges $1.25 and $4.25 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Does Gemma 3 27B or Muse Spark 1.3 have a bigger context window?
Muse Spark 1.3 has the larger context window: 131K for Gemma 3 27B against 1.0M for Muse Spark 1.3. Note that a window you can fill is not a window you should fill — retrieval quality usually degrades well before the limit, and you pay for every token you put in it.
Do Gemma 3 27B and Muse Spark 1.3 support prompt caching?
Both publish a cached-input rate: $0.040 per million for Gemma 3 27B and $0.150 for Muse Spark 1.3, against full input rates of $0.080 and $1.25. 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.
- 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.