Updated October 17, 2026. Rank says GPT-5.2 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
GPT-5.2 has the higher public score estimate, 59.86 versus 49.14, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 6 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
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.
Code generation, repair, and software-engineering tasks
GPT-5.2
GPT-5.2 has the higher public coding point estimate, 36.8 to 22.3, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Prompts that approach the documented context limit
Gemini 2.5 Pro
Gemini 2.5 Pro has the larger documented context window.
1K fresh input + 500 output tokens
Gemini 2.5 Pro
Gemini 2.5 Pro has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
Gemini 2.5 Pro
Gemini 2.5 Pro has the lower estimated token cost for this stated workload. GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
50K fresh input + 3K output tokens
Gemini 2.5 Pro
Gemini 2.5 Pro has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Gemini 2.5 Pro is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
Like-for-like · BenchAlign v5.8
GPT-5.2 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
FrontierMath v2 (Tiers 1-3)Math
Normalized gap 42.8FrontierMath v2 (Tier 4)Math
Normalized gap 31.7SWE-bench VerifiedCoding
Normalized gap 16.2GPQAKnowledge
Normalized gap 9.4Each row shows the public-lane category score for both models: the BenchAlign v5.8 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.
| Category | Gemini 2.5 Pro | GPT-5.2 | Basis | Reading |
|---|---|---|---|---|
| Coding | 22.3Supported · #115/150 | 36.8Supported · #73/150 | Like-for-likeBenchAlign v5.8 lane · 3 vs 3 public rows | GPT-5.2 leads · intervals overlap |
| Knowledge | 45.6Supported · #96/184 | 59.7Supported · #50/184 | Like-for-likeBenchAlign v5.8 lane · 2 vs 1 public rows | GPT-5.2 leads · intervals overlap |
| Agentic | 25.7Estimated · #102/128 | 44.4Supported · #58/128 | Directional onlyBenchAlign v5.8 lane · 1 vs 4 public rows | Directional only |
| Instruction following | 57.8#83/148 | 92.6#19/148 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 73.3Unranked · 2 rankable rows | 64.3Unranked · 4 rankable rows | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | 74.8Unranked · 1 rankable row | 75.9#22/54 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 36.9Unranked · 3 rankable rows | 71.1Unranked · 3 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
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.
1K fresh input + 500 output tokens
Gemini 2.5 Pro has the lower modeled cost
Costs use the listed standard API rates.
50K fresh input + 3K output tokens
Gemini 2.5 Pro has the lower modeled cost
Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
Gemini 2.5 Pro has the lower modeled cost
GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
Maximum documented context; output-token limits may be lower.
Gemini 2.5 Pro
GPT-5.2
400K
Gemini 2.5 Pro
gemini-2.5-pro
Google Gemini API pricingGPT-5.2
Not sourced
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Gemini 2.5 Pro
$0.125 per 1M cached input tokens
Google Gemini API pricingGPT-5.2
Not published
Gemini 2.5 Pro
Not sourced
GPT-5.2
Not sourced
Gemini 2.5 Pro
Not sourced
GPT-5.2
Not sourced
Gemini 2.5 Pro
Not sourced
GPT-5.2
Not sourced
Gemini 2.5 Pro
Non-Reasoning
GPT-5.2
Reasoning
Gemini 2.5 Pro
Proprietary
GPT-5.2
Proprietary
Gemini 2.5 Pro
Proprietary
GPT-5.2
Proprietary
Gemini 2.5 Pro
2025-03-01
GPT-5.2
2025-12-11
GPT-5.2 has the higher public score estimate, 59.86 versus 49.14, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
GPT-5.2 has the higher public coding point estimate, 36.8 to 22.3, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
GPT-5.2 scores higher for agentic tasks on the public lane, 44.4 to 25.7. Gemini 2.5 Pro 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.
For the stated presets, chat costs $0.00625 on Gemini 2.5 Pro and $0.00875 on GPT-5.2; repository review costs $0.0925 and $0.1295; the cache-heavy agent loop costs $0.15 and $0.525. GPT-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Gemini 2.5 Pro has the larger documented context window: 1M, compared with 400K.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Gert Labs
Shared sourceGPT-5.2 leads this result
BrowseComp
Not directly comparable
OSWorld-Verified
Not directly comparable
JobBench
Not directly comparable
SWE-bench Verified
GPT-5.2 leads this result
Vibe Code Bench
Shared sourceGPT-5.2 leads this result
SWE-bench (Vals)
Not directly comparable
SWE-bench Pro
Not directly comparable
MMMU
Not directly comparable
MMMU-Pro
Not directly comparable
RealWorldQA
Not directly comparable
OmniDocBench 1.5
Not directly comparable
Video-MME (with subtitle)
Not directly comparable
Video-MME (w/o subtitle)
Not directly comparable
MathVision
Not directly comparable
We-Math
Not directly comparable
DynaMath
Not directly comparable
SimpleVQA
Not directly comparable
CC-OCR
Not directly comparable
AI2D_TEST
Not directly comparable
CountBench
Not directly comparable
ERQA
Not directly comparable
VideoMMMU
Not directly comparable
MLVU (M-Avg)
Not directly comparable
CharXiv
Not directly comparable
V*
Not directly comparable
AIME 2024
Not directly comparable
FrontierMath v2 (Tiers 1-3)
Shared sourceGPT-5.2 leads this result
FrontierMath v2 (Tier 4)
Shared sourceGPT-5.2 leads this result
AIME 2025
Not directly comparable
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 5,600+ readers.
Last updated October 17, 2026
