LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

GPT-5.6 Luna

openai·openai/gpt-5.6-luna·GA·Closed·gpt-luna series
Cost-efficient GPT-5.6 model for fast, high-volume workloads

Specs & pricing

Input / output per 1M tokens
Official price·OpenAI
$0.20 / $1.20
Blended $0.45 · Cache read $0.02
Lowest paid·xxapiGateway
$0.012 / $0.074
Blended $0.028 · 81.2× spread
1 more $0 channels
Context
1,050,000
Output limit
128,000
Knowledge cutoff
2026-02-16
Released / updated
2026-07-09 / 2026-07-09
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF

GPT-5.6 Luna by openai is offered by 75 providers on this page. Public prices are shown for 73 of them. Its official list price is $0.20 per 1M input tokens and $1.20 per 1M output tokens. The lowest paid channel is xxapi at $0.012 / $0.074 per 1M, about 16× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.

Artificial Analysis rates it 23.8 on the Intelligence Index, with 40.7 for Strategy & Ops, its strongest domain. Against its lowest blended price of $0.028 per 1M, that is roughly 859 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 136 tokens per second, with 110.81s to the first token. Running one task of the Artificial Analysis suite costs about $0.3313, which reflects how many tokens its reasoning consumes rather than the unit price alone.

The context window is 1,050,000 tokens at the reference host, but hosts report different limits, from 128,000 to 1,050,000, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2026-02-16.

Quality & performanceArtificial Analysis · Intelligence Index v5 · rep. tier Max

Intelligence23.8
Value859
Engineering29.4
Finance & Accounting31.8
Legal27.7
Healthcare & Medical32.4
Strategy & Ops40.7
Economics32.8
Output speed136 tok/s
TTFT110.81 s
Cost per task$0.33
Value formulaIQ 23.8 ÷ min blended $0.028 = 859

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 75 providers73 with public prices · 1 free

  • ProviderInput / Output
  • GitLab DuoCloud
    duo-chat-gpt-5-6-luna
    Free
    Cache read $0Context 1,050,000Output limit 128,000
  • xxapiGateway
    $0.012 / $0.074
    tiered >272K: $0.025
    Cache read $0.0012Context 1,050,000Output limit 128,000
  • Outer LabsGateway
    $0.12 / $0.72
    tiered >272K: $0.24
    Cache read $0.012Context 1,050,000Output limit 128,000
  • KenariGateway
    gpt-5-6-luna
    $0.20 / $1.14
    Cache read $0.02Context 1,050,000Output limit 128,000
  • OpenAIOfficialFirst-party
    $0.20 / $1.20
    tiered >272K: $0.40
    Cache read $0.02Context 1,050,000Output limit 128,000
  • Perplexity AgentGateway
    $0.20 / $1.20
    tiered >272K: $0.40
    Cache read $0.02Context 1,050,000Output limit 128,000
  • CrossModelGateway
    $0.20 / $1.20
    tiered >272K: $0.40
    Cache read $0.02Context 1,050,000Output limit 128,000
  • BoundlessGateway
    $0.20 / $1.20
    Cache read $0.02Context 262,144 ⚠Output limit 128,000deprecated
  • LLM GatewayGateway
    $0.20 / $1.20
    Cache read $0.02Context 1,050,000Output limit 128,000
  • LLM GatewayGateway
    $0.20 / $1.20
    Cache read $0.02Context 1,050,000Output limit 128,000
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
GitLab Duo
duo-chat-gpt-5-6-luna
CloudFree$0$01,050,000128,000
xxapiGateway$0.012
tiered >272K: $0.025
$0.074$0.0012$0.0151,050,000128,000
Outer LabsGateway$0.12
tiered >272K: $0.24
$0.72$0.012—1,050,000128,000
Kenari
gpt-5-6-luna
Gateway$0.20$1.14$0.02$0.261,050,000128,000
OpenAIOfficialFirst-party$0.20
tiered >272K: $0.40
$1.20$0.02$0.251,050,000128,000
Perplexity AgentGateway$0.20
tiered >272K: $0.40
$1.20$0.02$0.251,050,000128,000
CrossModelGateway$0.20
tiered >272K: $0.40
$1.20$0.02$0.251,050,000128,000
BoundlessGateway$0.20$1.20$0.02—262,144 ⚠128,000deprecated
LLM GatewayGateway$0.20$1.20$0.02$0.251,050,000128,000
LLM GatewayGateway$0.20$1.20$0.02$0.251,050,000128,000

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

effort = noneeffort = loweffort = mediumeffort = higheffort = xhigheffort = maxeffort = minimal

Interleaved thinking (reasoning between tool calls) is declared by 6 of 75 providers.

2 / 75 providers expose no reasoning control (reasoning_options: []).

Experimental modes4 items

ModeProviderInputOutputCache readCache write
fastOpenAI$0.40$2.40$0.04$0.50
flexOpenAI$0.10$0.60$0.01$0.13
proOpenAI————
fastDatabricks$0.40$2.40$0.04$0.50

Your usage cost

Common scenarios
1xxapi$4.83
2Outer Labs$47.04
3Kenari$75.40
4OpenAI · Official$78.40
5Perplexity Agent$78.40
6CrossModel$78.40

1 more channels offer $0 (GitLab Duo); free tiers usually have rate limits and no SLA, excluded from ranking.

Switch to xxapi to save $73.57/mo (94%)
Note: this is a gateway; verify availability and rate limits yourself.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$1.20$02026-08-052026-11-15
2026-11-15Input list $0.20Output list $1.20Min blended $0.028

Benchmark12 items

NameConditionsScoreMetricSource
SWE-Bench Pro—62.7resolve rateSource ↗
Terminal-Benchv2.184.7success rateSource ↗
DeepSWEv1.167.2resolve rateSource ↗
GPQA Diamond—92.3accuracySource ↗
FrontierMathdataset: Tier 1-3 · vv278.6accuracySource ↗
BrowseComp—83.3accuracySource ↗
OSWorldv2.045.6success rateSource ↗
MMMU Provariant: no tools78.4accuracySource ↗
Agents' Last Exam—50.3—Source ↗
Toolathlon—53.4success rateSource ↗
Artificial Analysis Intelligence Indexvariant: max · v4.151.2index scoreSource ↗
Artificial Analysis Coding Agent Indexharness: Codex · variant: max · v1.174.6index scoreSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Related models

GPT-6 Lunasame series$0.10 / $0.50GPT Luna Latestsame series—GLM-5.3-Flashcheaper alternative$0.15 / $0.50DeepSeek V4.1 Flash 1017cheaper alternative$0.15 / $0.60Claude Haiku 5.5cheaper alternative$0.10 / $0.50GPT-5.5 Prosame lab$30.00 / $180GPT-5.6 Solsame lab$4.00 / $20.00
Data partly from models.dev (MIT) · OpenAI official docs ↗