Agentic
Directional only- GPT-5.6 Terra
- 60.2
- Supported · #20/152
- Mercury 2.5
- 44.5
- Estimated · #91/152
- Basis
- BenchAlign lane · 8 vs 2 public rows
- Reading
- Directional only
GPT-5.6 Terra vs Mercury 2.5
Updated September 8, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Long documents
Prompts that approach the documented context limit
GPT-5.6 Terra
GPT-5.6 Terra has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
Mercury 2.5
Mercury 2.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Mercury 2.5
Mercury 2.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Show secondary and unsupported calls
Repository review cost
50K fresh input + 3K output tokens
Mercury 2.5
Mercury 2.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Mercury 2.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Mercury 2.5 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
30 GPT-5.6 Terra only
1 Shared
4 Mercury 2.5 only
- Shared results
- 1
- GPT-5.6 Terra only
- 30
- Mercury 2.5 only
- 4
- Like-for-like categories
- 0 / 8
4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Coding
Directional only- GPT-5.6 Terra
- 67.1
- Supported · #9/151
- Mercury 2.5
- 47.8
- Estimated · #73/151
- Basis
- BenchAlign lane · 8 vs 1 public rows
- Reading
- Directional only
Knowledge
Directional only- GPT-5.6 Terra
- 69.3
- Supported · #15/182
- Mercury 2.5
- 48.3
- Estimated · #94/182
- Basis
- BenchAlign lane · 8 vs 1 public rows
- Reading
- Directional only
Instruction following
Directional only- GPT-5.6 Terra
- 87.1
- #34/121
- Mercury 2.5
- 80.1
- #53/121
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- GPT-5.6 Terra
- 64.0
- #11/18
- Mercury 2.5
- 67.7
- Unranked · 1 rankable row
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.6 Terra
- 97.0
- Unranked · 3 rankable rows
- Mercury 2.5
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.6 Terra
- Not ranked
- Mercury 2.5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.6 Terra
- 77.1
- #15/48
- Mercury 2.5
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
| Category | GPT-5.6 Terra | Mercury 2.5 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 60.2Supported · #20/152 | 44.5Estimated · #91/152 | Directional onlyBenchAlign lane · 8 vs 2 public rows | Directional only |
| Coding | 67.1Supported · #9/151 | 47.8Estimated · #73/151 | Directional onlyBenchAlign lane · 8 vs 1 public rows | Directional only |
| Knowledge | 69.3Supported · #15/182 | 48.3Estimated · #94/182 | Directional onlyBenchAlign lane · 8 vs 1 public rows | Directional only |
| Instruction following | 87.1#34/121 | 80.1#53/121 | Directional onlyProvisional lane · 0 vs 1 weighted rows | Directional only |
| Reasoning | 64.0#11/18 | 67.7Unranked · 1 rankable row | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Math | 97.0Unranked · 3 rankable rows | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 77.1#15/48 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- GPT-5.6 Terra
- $0.01
- Fits in one request
- Mercury 2.5
- $0.00012
- Fits in one request
Mercury 2.5 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- GPT-5.6 Terra
- $0.17
- Fits in one request
- Mercury 2.5
- $0.00245
- Fits in one request
Mercury 2.5 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- GPT-5.6 Terra
- $0.25
- Fits in one request
- Mercury 2.5
- $0.0031
- Fits in one request
Mercury 2.5 has the lower modeled cost
Costs use the listed standard API rates.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGPT-5.6 TerraMercury 2.5
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.6 Terra
1.05M
OpenAI model catalogMercury 2.5
API model ID
GPT-5.6 Terra
gpt-5.6-terra
OpenAI model catalogMercury 2.5
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.6 Terra
$0.25 per 1M cached input tokens
OpenAI API pricingMercury 2.5
$0.004 per 1M cached input tokens
Inception models and Mercury 2.5 launch pricingDocumented inputs
GPT-5.6 Terra
text, image
OpenAI model catalogMercury 2.5
Not sourced
Documented outputs
GPT-5.6 Terra
Mercury 2.5
Not sourced
Provider availability
GPT-5.6 Terra
Generally Available · OpenAI Responses API
OpenAI model catalogMercury 2.5
Not sourced
Reasoning profile
GPT-5.6 Terra
Reasoning
Mercury 2.5
Reasoning
Weight access
GPT-5.6 Terra
Proprietary
Mercury 2.5
Proprietary
License
GPT-5.6 Terra
Proprietary
Mercury 2.5
Proprietary
Release date
GPT-5.6 Terra
2026-07-09
Mercury 2.5
2026-09-08
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- Repository review: $0.17 vs $0.00245. Cache-heavy agent loop: $0.25 vs $0.0031.
- Context tradeoff
- GPT-5.6 Terra has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence35 rows
Agentic
Terminal-Bench 3.0
GPT-5.6 Terra20.8%SourceMercury 2.5—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Terra87.4%SourceMercury 2.5—Not directly comparable
BrowseComp
GPT-5.6 Terra87.5%SourceMercury 2.5—Not directly comparable
OSWorld 2.0
GPT-5.6 Terra50.2%SourceMercury 2.5—Not directly comparable
CyberGym
GPT-5.6 Terra81.8%SourceMercury 2.5—Not directly comparable
ExploitGym
GPT-5.6 Terra23.2%SourceMercury 2.5—Not directly comparable
Toolathlon
GPT-5.6 Terra53.1%SourceMercury 2.5—Not directly comparable
Terminal-Bench 2.1 (Vals)
GPT-5.6 Terra77.5%SourceMercury 2.5—Not directly comparable
τ³-bench results
GPT-5.6 Terra—Mercury 2.596.0%SourceNot directly comparable
DeepSearchQA
GPT-5.6 Terra—Mercury 2.534.0%SourceNot directly comparable
Coding
SWE-bench Pro
GPT-5.6 Terra63.4%SourceMercury 2.5—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Terra87.4%SourceMercury 2.5—Not directly comparable
deepSwe
GPT-5.6 Terra69.6%SourceMercury 2.5—Not directly comparable
FrontierCode 1.1 Extended
GPT-5.6 Terra55.8%SourceMercury 2.5—Not directly comparable
cursorBench32
GPT-5.6 Terra64.9%SourceMercury 2.5—Not directly comparable
VulcanBench v3
GPT-5.6 Terra87.0%SourceMercury 2.5—Not directly comparable
LiveCodeBench (Vals)
GPT-5.6 Terra85.9%SourceMercury 2.5—Not directly comparable
SWE-bench (Vals)
GPT-5.6 Terra95.4%SourceMercury 2.5—Not directly comparable
SciCode
GPT-5.6 Terra—Mercury 2.538%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GPT-5.6 Terra92.9%SourceMercury 2.5—Not directly comparable
GPQA-D
GPT-5.6 Terra92.9%SourceMercury 2.579.0%SourceGPT-5.6 Terra leads this result
HLE-Verified
GPT-5.6 Terra51.1%SourceMercury 2.5—Not directly comparable
LABBench2
GPT-5.6 Terra81.2%SourceMercury 2.5—Not directly comparable
HealthBench Professional
GPT-5.6 Terra57.7%SourceMercury 2.5—Not directly comparable
HealthBench Hard
GPT-5.6 Terra32.7%SourceMercury 2.5—Not directly comparable
GPQA Diamond (Vals)
GPT-5.6 Terra90.9%SourceMercury 2.5—Not directly comparable
MMLU-Pro (Vals)
GPT-5.6 Terra86.7%SourceMercury 2.5—Not directly comparable
Math
Multimodal
Instruction following
IFBench
GPT-5.6 Terra—Mercury 2.577%SourceNot directly comparable
Frequently asked questions
Which is better, GPT-5.6 Terra or Mercury 2.5?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, GPT-5.6 Terra or Mercury 2.5?
GPT-5.6 Terra scores higher for coding on the public lane, 67.1 to 47.8. Mercury 2.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, GPT-5.6 Terra or Mercury 2.5?
GPT-5.6 Terra scores higher for agentic tasks on the public lane, 60.2 to 44.5. Mercury 2.5 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, GPT-5.6 Terra or Mercury 2.5?
For the stated presets, chat costs $0.01 on GPT-5.6 Terra and $0.00012 on Mercury 2.5; repository review costs $0.17 and $0.00245; the cache-heavy agent loop costs $0.25 and $0.0031. Costs use the listed standard API rates.
Which has the larger context window, GPT-5.6 Terra or Mercury 2.5?
GPT-5.6 Terra has the larger documented context window: 1.05M, compared with 260K.
Related comparisons
Last updated September 8, 2026
Watch GPT-5.6 Terra vs Mercury 2.5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.