Capability
72.3/100
field median 57.6
#12 of 216 ranked models
Provider changes are easy to miss. Radar watches releases, pricing, deprecations, and incidents at the source.Provider changes are easy to miss.
See RadarModel profile · OpenAI
Released Jul 9, 2026 — see all recent releases
Data as of August 10, 2026 · How the score is built
Agentic ranks #10. Particularly useful for coding agents, browser research, and computer-use workflows.
23 published rows leave some tracked benchmark slots empty. Independent runtime speed has not been measured.
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
72.3/100
field median 57.6
#12 of 216 ranked models
Price
$2input / $12 output
input median $1
cached $0.20 · blended $7
Speed
Not measured
field median 89 tok/s
Time to first token not measured
Context
1.05Mtokens
field median 200,000
Maximum output length is tracked separately
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
GPT-5.6 Terra category percentile values
The dashed outline is median of 6 nearest peers.
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #10 of 133Percentile 93rdWeight 22%6 benchmarksVerified | 63.3 | #10 of 133 | 93rd | 22% | 6 benchmarks | Verified |
| CodingRank #11 of 133Percentile 92ndWeight 20%5 benchmarksVerified | 65.1 | #11 of 133 | 92nd | 20% | 5 benchmarks | Verified |
| ReasoningRank Not rankedWeight 17%2 benchmarksVerified | 85.9 | Not ranked | Not available | 17% | 2 benchmarks | Verified |
| KnowledgeRank #10 of 55Percentile 83rdWeight 12%4 benchmarksVerified | 81.9 | #10 of 55 | 83rd | 12% | 4 benchmarks | Verified |
| MathRank Not rankedWeight 5%3 benchmarksVerified | 96.9 | Not ranked | Not available | 5% | 3 benchmarks | Verified |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalRank #10 of 35Percentile 74thWeight 12%2 benchmarksVerified | 75.3 | #10 of 35 | 74th | 12% | 2 benchmarks | Verified |
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score63.4% | Versus best verified row Best verified: Claude Mythos 5 · 80.3% | Gap16.9 behind | WeightWeighted 10% | Provider exact |
| Terminal-Bench 2.0 | Score87.4% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap4.5 behind | WeightDisplay only | Provider exact |
| deepSwe | Score69.6% | Versus best verified row Best verified: GPT-5.6 Sol · 72.7% | Gap3.1 behind | WeightDisplay only | Provider exact |
| FrontierCode 1.1 Extended | Score55.8% | Versus best verified row Best verified: Claude Opus 5 · 63.6% | Gap7.8 behind | WeightDisplay only | Provider exact |
| cursorBench32 | Score64.9% | Versus best verified row Best verified: Claude Fable 5 · 70.5% | Gap5.6 behind | WeightDisplay only | Benchmark exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 | Score87.4% | Versus best verified row Best verified: GPT-5.6 Sol · 91.9% | Gap4.5 behind | WeightWeighted 38% | Provider exact |
| BrowseComp | Score87.5% | Versus best verified row Best verified: GPT-5.6 Sol · 92.2% | Gap4.7 behind | WeightWeighted 28% | Provider exact |
| OSWorld 2.0 | Score50.2% | Versus best verified row Best verified: Claude Opus 5 · 70.6% | Gap20.4 behind | WeightDisplay only | Provider exact |
| CyberGym | Score81.8% | Versus best verified row Best verified: Fugu Cyber · 86.9% | Gap5.1 behind | WeightDisplay only | Provider exact |
| ExploitGym | Score23.2% | Versus best verified row Best verified: GPT-5.6 Sol · 33.7% | Gap10.5 behind | WeightDisplay only | Provider exact |
| Toolathlon | Score53.1% | Versus best verified row Best verified: Muse Spark 1.1 · 75.6% | Gap22.5 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | Score83.9% | Versus best verified row Best verified: GPT-5.6 Sol · 92.5% | Gap8.6 behind | WeightWeighted 31% | Benchmark exact |
| ARC-AGI-3Abstraction and Reasoning Corpus for AGI v3 | Score0.8% | Versus best verified row Best verified: Claude Opus 5 · 30.2% | Gap29.4 behind | WeightDisplay only | Benchmark exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| GPQAGraduate-Level Google-Proof Q&A | Score92.9% | Versus best verified row Best verified: Sakana Fugu-Ultra · 95.5% | Gap2.6 behind | WeightWeighted 7% | Provider exact |
| GPQA-DGPQA Diamond | Score92.9% | Versus best verified row Best verified: Sakana Fugu-Ultra · 95.5% | Gap2.6 behind | WeightDisplay only | Provider exact |
| HealthBench Professional | Score57.7% | Versus best verified row Best verified: GPT-5.6 Sol · 60.5% | Gap2.8 behind | WeightDisplay only | Provider exact |
| HealthBench Hard | Score32.7% | Versus best verified row Best verified: Muse Spark · 42.8% | Gap10.1 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-3 | Score84.900% | Versus best verified row Best verified: GPT-5.6 Sol · 89.000% | Gap4.1 behind | WeightWeighted 30% | Provider exact |
| FrontierMath v2 (Tier 4)FrontierMath v2 Tier 4 | Score68.300% | Versus best verified row Best verified: GPT-5.6 Sol · 83.000% | Gap14.7 behind | WeightWeighted 10% | Provider exact |
| FrontierMath (legacy)FrontierMath legacy aggregate | Score84.9% | Versus best verified row Best verified: GPT-5.6 Sol · 89% | Gap4.1 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMMU-ProMassive Multi-discipline Multimodal Understanding Pro | Score80.7% | Versus best verified row Best verified: GPT-5.4 Pro · 94% | Gap13.3 behind | WeightWeighted 45% | Provider exact |
| MMMU-Pro w/ PythonMMMU-Pro with Python | Score82% | Versus best verified row Best verified: GPT-5.6 Sol · 84.6% | Gap2.6 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ExploitBenchExploitBench v8-bench | Score53% | Versus best verified row Best verified: GPT-5.6 Sol · 74% | Gap20.6 behind | WeightDisplay only | Provider exact |
The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Jul 9, 2026 · you are here
GPT-5.6 TerraScore 72.3 · $2 / $12
terra · terra
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
GPT-5.6 Terra ranks #12 of 216 on the public leaderboard with a score of 72.29/100. It does not yet have enough sourced coverage for a verified position.
GPT-5.6 Terra is a proprietary model with a 1.05M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
GPT-5.6 Terra sits in the GPT-5.6 family with GPT-5.6 Sol, GPT-5.6 Luna. 23 of 381 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Agentic at #10, while its lowest eligible position is Coding at #11. particularly useful for coding agents, browser research, and computer-use workflows.
OpenAI · Model release
GPT-5.6 Terra ranks #12 out of 216 models on the public BenchAlign leaderboard, with a score of 72.29/100. Its evidence status is Estimated, and this profile shows 23 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
GPT-5.6 Terra ranks #10 out of 55 eligible models for knowledge and understanding, with a public category score of 81.9/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
GPT-5.6 Terra ranks #11 out of 133 eligible models for coding and programming, with a public category score of 65.1/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
GPT-5.6 Terra has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
GPT-5.6 Terra has source-displayable benchmark coverage for reasoning and logic, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
GPT-5.6 Terra ranks #10 out of 133 eligible models for agentic tool use and computer tasks, with a public category score of 63.3/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
GPT-5.6 Terra ranks #10 out of 35 eligible models for multimodal and grounded tasks, with a public category score of 75.3/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
GPT-5.6 Terra belongs to the GPT-5.6 family. Related tracked variants include GPT-5.6 Sol, GPT-5.6 Luna. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
No. GPT-5.6 Terra currently has 46 source-displayable rows across 381 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
GPT-5.6 Terra has a documented context window of 1.05M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.
Related resources
Last updated August 10, 2026. Runtime fields remain blank until a sourced snapshot exists.
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.