Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d66, 14 pendingSources1,015

GPT-5.6 Terra

OpenAIReleased Jul 9, 2026Current

Score basis

GPT-5.6 Terra is at the frontier in long context; strong in reasoning; capable in coding, agentic tasks, multimodal tasks, and math; and behind the leaders in factuality and instruction following. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Frontier1 capability
  1. Long ContextLeads2 of 3
Strong1 capability
  1. Reasoning−8.8%6 of 6
Capable4 capabilities
  1. Coding−12.1%7 of 10
  2. Agentic−15.6%4 of 7
  3. Multimodal−18.9%2 of 6
  4. Math−20.7%3 of 5
Limited2 capabilities
  1. Factuality−30.4%3 of 4
  2. Instruction Following−31.6%2 of 3
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$2.00input$12.00outputper million tokens

From Artificial Analysis · 2 providers tracked · All prices

Costs more than 85% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

176results on65benchmarks

  • 30 independently verified
  • 110 aggregator
  • 22 vendor-reported
  • 14 cross-referenced

From 16 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

8 papers reference GPT-5.6 Terra

GPT-5.6 Terra benchmark results

176 results on 65 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

FrontierFull long context ranking

Leads the field2 of 3 ranked benchmarks measured

Show 5 more long context resultsHide 5 long context results

Reasoning

StrongFull reasoning ranking

8.8% behind the leader6 of 6 ranked benchmarks measured

Show 26 more reasoning resultsHide 26 reasoning results

Coding

CapableFull coding ranking

12.1% behind the leader7 of 10 ranked benchmarks measured

Show 14 more coding resultsHide 14 coding results

Agentic

CapableFull agentic ranking

15.6% behind the leader4 of 7 ranked benchmarks measured

Show 16 more agentic resultsHide 16 agentic results

Multimodal

CapableFull multimodal ranking

18.9% behind the leader2 of 6 ranked benchmarks measured

Show 7 more multimodal resultsHide 7 multimodal results

Math

CapableFull math ranking

20.7% behind the leader3 of 5 ranked benchmarks measured

Show 1 more math resultHide 1 math result

Factuality

LimitedFull factuality ranking

30.4% behind the leader3 of 4 ranked benchmarks measured

Show 10 more factuality resultsHide 10 factuality results

Instruction Following

LimitedFull instruction following ranking

31.6% behind the leader2 of 3 ranked benchmarks measured

Show 4 more instruction following resultsHide 4 instruction following results

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 58 more resultsHide 58 results

GPT-5.6 Terra: common questions

Who makes GPT-5.6 Terra?

GPT-5.6 Terra is made by OpenAI.

When was GPT-5.6 Terra released?

GPT-5.6 Terra was released on Jul 9, 2026, according to Artificial Analysis.

What is GPT-5.6 Terra good at?

GPT-5.6 Terra is at the frontier in long context; strong in reasoning; capable in coding, agentic tasks, multimodal tasks, and math; and behind the leaders in factuality and instruction following. Too few results yet to rate safety or multilingual tasks.

How much does GPT-5.6 Terra cost?

GPT-5.6 Terra costs $2.00 per million input tokens and $12.00 per million output tokens, according to Artificial Analysis. We track its price at 2 providers. At a mix of three input tokens to one output token, it costs more than 85% of the 330 priced models we track.

How many benchmarks has GPT-5.6 Terra been tested on?

We track 176 results for GPT-5.6 Terra on 65 benchmarks from 16 sources, 30 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does GPT-5.6 Terra support?

OpenRouter lists tool calling, structured outputs, and reasoning for GPT-5.6 Terra.

About this record

Where GPT-5.6 Terra's numbers come from, and every name it appears under.

Tracked since
Jul 1, 2026
Newest source mention
Sep 9, 2026

Where the results come from

Verification: 176 scores · 30 independently verified · 110 aggregator-attributed · 14 vendor cross-reference · 22 vendor-reported. How these tiers are assigned

From 16 sources on 10 sites. Artificial Analysis supplies 111 of them; the 30 independently verified results come from 7 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai111
  • api.llm-stats.com18
  • arcprize.org15
  • deepmind.google14
  • livebench.ai7
  • deploymentsafety.openai.com4
  • epoch.ai3
  • datasets-server.huggingface.co2
  • lmarena.ai1
  • simple-bench.com1

Also known as

How our sources name GPT-5.6 Terra at each reasoning setting.

SettingShort formAPI id
lowgpt-5.6 terra (low)gpt-5-6-terra-low
mediumgpt-5.6 terra (medium)gpt-5-6-terra-medium
highgpt-5.6 terra (high)gpt-5-6-terra-high
xhighgpt-5.6 terra (xhigh)gpt-5-6-terra-xhigh gpt-5.6-terra-xhigh gpt-5.6-terra-xhigh (codex-harness)
maxgpt-5.6 terra (max) gpt-5.6 terra max—
Also listed asgpt-5-6-terragpt-5-6-terra-non-reasoninggpt-5.6 terra (non-reasoning)gpt-5.6-terra:batchchatgpt 5.6-terra