// head_to_head
Codestral 2508 vs GPT-5.6 Luna
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.
Codestral 2508 and GPT-5.6 Luna are priced within ~10% of each other.
Codestral 2508 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.
mistralai
Codestral 2508
- Blended / 1M
- $0.450
- Context
- 256K
- Released
- Aug 1, 2025
- Overall score
- Not evaluated
openai
GPT-5.6 Luna
- Blended / 1M
- $0.450
- Context
- 1.1M
- Released
- Jul 9, 2026
- Overall score
- 73.6
Specs and pricing
| Metric | Codestral 2508 | GPT-5.6 Luna |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 73.6 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.0911 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.450 | $0.450 |
| Input price / 1M | $0.300 | $0.200win |
| Output price / 1M | $0.900win | $1.20 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.030 | $0.020win |
| Context window | 256K | 1.1Mwin |
| Max output tokens | 205Kwin | 128K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Codestral 2508 on top, GPT-5.6 Luna 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 | Codestral 2508 | GPT-5.6 Luna |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $124.56/mo | $131.04/mo |
| RAG assistant 8K in / 600 out × 100K requests | $186.00/mo | $160.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $160.80/mo | $155.20/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $354.00/mo | $281.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $705.00/mo | $530.00/mo |
Which should you pick?
You need to fit large documents in one call
GPT-5.6 Luna
Wider context window — 1.1M against 256K.
Codestral 2508 vs GPT-5.6 Luna FAQ
Which is better, Codestral 2508 or GPT-5.6 Luna?
Codestral 2508 and GPT-5.6 Luna are priced within ~10% of each other. Codestral 2508 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 Codestral 2508 cheaper than GPT-5.6 Luna?
They cost about the same. Both land near $0.450 per million tokens on a 3:1 input:output blend, so price is unlikely to be the deciding factor between them.
Does Codestral 2508 or GPT-5.6 Luna have a bigger context window?
GPT-5.6 Luna has the larger context window: 256K for Codestral 2508 against 1.1M for GPT-5.6 Luna. 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 Codestral 2508 and GPT-5.6 Luna support prompt caching?
Both publish a cached-input rate: $0.030 per million for Codestral 2508 and $0.020 for GPT-5.6 Luna, against full input rates of $0.300 and $0.200. 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.