// head_to_head
Kimi K2.6 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.
Qwen3.8 27B scores higher, Kimi K2.6 costs less — it depends on your workload.
Qwen3.8 27B is ahead by 4.7 points overall, and Kimi K2.6 lists 13% cheaper per blended million tokens. Whether 4.7 points is worth that depends on how much a wrong answer costs you. The two cost measures disagree here, which is worth knowing: Kimi K2.6 has the lower sticker price, but Qwen3.8 27B earns each point of capability for less — $0.0556 against $0.0918 — because per-token rates do not predict how many tokens a model actually spends on a task.
moonshotai
Kimi K2.6
- Blended / 1M
- $1.01
- Context
- 262K
- Released
- Apr 20, 2026
- Overall score
- 70.5
qwen
Qwen3.8 27B
- Blended / 1M
- $1.14
- Context
- 1M
- Released
- Aug 14, 2026
- Overall score
- 75.3
Specs and pricing
| Metric | Kimi K2.6 | Qwen3.8 27B |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | 70.5 | 75.3win |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.0918 | $0.0556win |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $1.01win | $1.14 |
| Input price / 1M | $0.560 | $0.450win |
| Output price / 1M | $2.36win | $3.20 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.094 | $0.050win |
| Context window | 262K | 1Mwin |
| Max output tokens | 262Kwin | 131K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Kimi K2.6 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 | Kimi K2.6 | Qwen3.8 27B |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $289.76/mo | $335.20/mo |
| RAG assistant 8K in / 600 out × 100K requests | $403.56/mo | $392.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $376.18/mo | $392.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $714.19/mo | $670.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $1404.20/mo | $1245.00/mo |
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 46.9.
You need to fit large documents in one call
Qwen3.8 27B
Wider context window — 1M against 262K.
You are cost-constrained
Kimi K2.6
Cheaper on blended list price at $1.01 per million tokens.
Kimi K2.6 vs Qwen3.8 27B FAQ
Which is better, Kimi K2.6 or Qwen3.8 27B?
Qwen3.8 27B scores higher, Kimi K2.6 costs less — it depends on your workload. Qwen3.8 27B is ahead by 4.7 points overall, and Kimi K2.6 lists 13% cheaper per blended million tokens. Whether 4.7 points is worth that depends on how much a wrong answer costs you. The two cost measures disagree here, which is worth knowing: Kimi K2.6 has the lower sticker price, but Qwen3.8 27B earns each point of capability for less — $0.0556 against $0.0918 — because per-token rates do not predict how many tokens a model actually spends on a task.
Is Kimi K2.6 cheaper than Qwen3.8 27B?
Kimi K2.6 is cheaper. On a 3:1 input:output blend, Kimi K2.6 lists at $1.01 per million tokens and Qwen3.8 27B at $1.14 — Kimi K2.6 is 13% cheaper. Input and output are priced separately — Kimi K2.6 charges $0.560 in and $2.36 out, Qwen3.8 27B charges $0.450 and $3.20 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Kimi K2.6 vs Qwen3.8 27B: which scores higher on benchmarks?
Kimi K2.6 scores 70.5 and Qwen3.8 27B scores 75.3 overall on LiveBench, the mean of its seven categories. That is a 4.7-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, Kimi K2.6 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. Kimi K2.6 works out at $0.0918 per point and Qwen3.8 27B at $0.0556.
Does Kimi K2.6 or Qwen3.8 27B have a bigger context window?
Qwen3.8 27B has the larger context window: 262K for Kimi K2.6 against 1M for Qwen3.8 27B. 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 Kimi K2.6 and Qwen3.8 27B support prompt caching?
Both publish a cached-input rate: $0.094 per million for Kimi K2.6 and $0.050 for Qwen3.8 27B, against full input rates of $0.560 and $0.450. 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.