GPT-5.6 Luna
LiveOpenAI1.1M context · Text, Vision, Tools, Reasoning
Best discount now98.2%offYou pay 1.8% of reference.
Median discount80.0%offAcross 77 live offers.
Realized · 7 daysNot enough dataIt fills in after the nightly update.
Requests · 24h236,13477 offers from 44 sellers.
Current offers
77 of 938 offers
Availability
Providers
ProviderDiscountPriceLatencyUptimeOfferedSoldRemaining
InferHub ?
98%
$0.00357 in
$0.0214 out
10.1s
75.4%
$1B
$0.381
$1B
InferHub ?
98%
$0.00394 in
$0.0236 out
10.1s
75.4%
∞
-
∞
InferHub ?
98%
$0.00398 in
$0.0239 out
10.1s
75.4%
$115
$0.051
$115
Amazon Bedrock
98%
$0.00398 in
$0.0239 out
-
-
∞
$302
∞
Amazon Bedrock
98%
$0.00398 in
$0.0239 out
-
-
∞
$35
∞
InferHub ?
97%
$0.00427 in
$0.0256 out
10.1s
75.4%
$1,000
-
$1,000
InferHub ?
95%
$0.00871 in
$0.0523 out
10.1s
75.4%
∞
$7
∞
InferHub ?
95%
$0.00871 in
$0.0523 out
10.1s
75.4%
$888
$0.003
$888
InferHub ?
95%
$0.00871 in
$0.0523 out
10.1s
75.4%
$888
-
$888
InferHub ?
95%
$0.00871 in
$0.0523 out
10.1s
75.4%
∞
$0.181
∞
Total
Cumulative offers
Offered
∞
Sold
$486
Remaining
∞
Discount trend
Realized · 7 daysNo trend yet.It fills in after the nightly update.
Recent fills
17:1021,034 tokensInferHub98.3%
17:1053,086 tokensInferHub98.3%
17:1055,699 tokensOpenAI direct95.5%
17:1018,427 tokensInferHub98.3%
17:101,727 tokensOpenRouter60.0%
17:102,275 tokensAWS Bedrock98.0%
17:102,600 tokensInferHub98.4%
17:1013,083 tokensOpenAI direct95.5%
Settled on Base in USDC.
Performance by provider route · 24h
Measured on Surplus traffic| Route | p50 latency | p95 latency | Throughput | Success rate | ShareShare of traffic |
|---|---|---|---|---|---|
| OpenAI direct | 4.11 s | 18.42 s | 99 tok/s | 97.2% | 63% |
| InferHub | 13.64 s | 82.59 s | 243 tok/s | 75.4% | 35% |
| OpenRouter | 5.55 s | 10.11 s | — | 89.9% | 1% |
| Venice AI direct | 1.67 s | 7.49 s | — | 51.6% | 1% |
OpenAI direct63% of traffic
97.2% success
InferHub35% of traffic
75.4% success
OpenRouter1% of traffic
89.9% success
Venice AI direct1% of traffic
51.6% success
Use this modelChange your base URL.
Change your base URL.
Keep your code.
Any OpenAI-compatible client works. Add /min30 to the path to only buy at 30% off or better.
Read the quickstart ↗curl $SURPLUS_BASE_URL/chat/completions \ -H "Authorization: Bearer $SURPLUS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "gpt-5.6-luna", "messages": [{"role": "user", "content": "Hello"}] }'
SURPLUS_BASE_URL
https://api.surplusintelligence.ai/v1