// head_to_head
Gemini 4 Carbon vs Qwen3.8 27B
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.
Gemini 4 Carbon scores higher, Qwen3.8 27B costs less — it depends on your workload.
Gemini 4 Carbon is ahead by 29.5 points overall, and Qwen3.8 27B lists 6.9× cheaper per blended million tokens. Whether 29.5 points is worth that depends on how much a wrong answer costs you. Qwen3.8 27B also leads on measured cost per point of capability, at $0.0106 per point.
Gemini 4 Carbon
- Blended / 1M
- $4.00
- Context
- 1.0M
- Released
- Nov 10, 2026
- Overall score
- 83.5
qwen
Qwen3.8 27B
- Blended / 1M
- $0.581
- Context
- 1.0M
- Released
- Aug 14, 2026
- Overall score
- 54.0
Specs and pricing
| Metric | Gemini 4 Carbon | Qwen3.8 27B |
|---|---|---|
| LiveBench overall Weighted mean of the seven LiveBench category scores, 0–100. Higher is better. | 83.5win | 54.0 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.0202 | $0.0106win |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $4.00 | $0.581win |
| Input price / 1M | $2.00 | $0.150win |
| Output price / 1M | $10.00 | $1.88win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.100 | $0.037win |
| Context window | 1.0M | 1.0M |
| Max output tokens | 944Kwin | 236K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Gemini 4 Carbon on top, Qwen3.8 27B 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 | Gemini 4 Carbon | Qwen3.8 27B |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $1,143/mo | $177.90/mo |
| RAG assistant 8K in / 600 out × 100K requests | $1,440/mo | $187.50/mo |
| Coding agent 40K in / 4K out × 20K requests | $1,336/mo | $207.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $2,655/mo | $285.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $5,050/mo | $506.25/mo |
Which should you pick?
You are running this at volume
Qwen3.8 27B
Lowest measured cost per point of capability at $0.0106 per point — the gap compounds with every request.
Quality matters more than the bill
Gemini 4 Carbon
Highest overall LiveBench score of the two at 83.5.
The workload is coding or agentic work
Gemini 4 Carbon
Leads on long-horizon agentic — 72.7 against 15.9.
Gemini 4 Carbon vs Qwen3.8 27B FAQ
Which is better, Gemini 4 Carbon or Qwen3.8 27B?
Gemini 4 Carbon scores higher, Qwen3.8 27B costs less — it depends on your workload. Gemini 4 Carbon is ahead by 29.5 points overall, and Qwen3.8 27B lists 6.9× cheaper per blended million tokens. Whether 29.5 points is worth that depends on how much a wrong answer costs you. Qwen3.8 27B also leads on measured cost per point of capability, at $0.0106 per point.
Is Gemini 4 Carbon cheaper than Qwen3.8 27B?
Qwen3.8 27B is cheaper. On a 3:1 input:output blend, Gemini 4 Carbon lists at $4.00 per million tokens and Qwen3.8 27B at $0.581 — Qwen3.8 27B is 6.9× cheaper. Input and output are priced separately — Gemini 4 Carbon charges $2.00 in and $10.00 out, Qwen3.8 27B charges $0.150 and $1.88 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Gemini 4 Carbon vs Qwen3.8 27B: which scores higher on benchmarks?
Gemini 4 Carbon scores 83.5 and Qwen3.8 27B scores 54.0 overall on LiveBench, the weighted mean of its seven categories. That is a 29.5-point lead for Gemini 4 Carbon. 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, Gemini 4 Carbon or Qwen3.8 27B?
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. Gemini 4 Carbon works out at $0.0202 per point and Qwen3.8 27B at $0.0106.
Does Gemini 4 Carbon or Qwen3.8 27B have a bigger context window?
They are effectively the same — 1.0M for Gemini 4 Carbon and 1.0M for Qwen3.8 27B.
Do Gemini 4 Carbon and Qwen3.8 27B support prompt caching?
Both publish a cached-input rate: $0.100 per million for Gemini 4 Carbon and $0.037 for Qwen3.8 27B, against full input rates of $2.00 and $0.150. 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 2026-11-06, 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.