Updated October 28, 2026. Rank says Gemini 4 Carbon is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Gemini 4 Carbon has the higher public point estimate, 88.42 versus 68.83. Their conditional score ranges overlap. These ranges do not establish rank confidence. 1 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
Gemini 4 Carbon
Gemini 4 Carbon has the higher public coding point estimate, 82.2 to 62.4, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
1K fresh input + 500 output tokens
Gemini 4 Carbon
Gemini 4 Carbon 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 4 Carbon
Gemini 4 Carbon has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
50K fresh input + 3K output tokens
Gemini 4 Carbon
Gemini 4 Carbon 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 4 Carbon is not ranked on the public lane for agentic, so no winner is named for agentic.
Prompts that approach the documented context limit
Not enough matched evidence
A complete context comparison is not sourced.
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
Gemini 4 Carbon 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.
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.
Each 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 4 Carbon | GPT-5.5 | Basis | Reading |
|---|---|---|---|---|
| Coding | 82.2Supported · #3/156 | 62.4Supported · #15/156 | Like-for-likeBenchAlign v5.8 lane · 1 vs 11 public rows | Gemini 4 Carbon leads |
| Knowledge | 88.2Supported · #2/178 | 71.2Supported · #14/178 | Like-for-likeBenchAlign v5.8 lane · 0 vs 6 public rows | Gemini 4 Carbon leads · intervals overlap |
| Agentic | Not ranked | 61.0Supported · #25/134 | Not comparableBenchAlign v5.8 lane · 4 vs 15 public rows | Not comparable |
| Reasoning | 79.5Unranked · 2 rankable rows | 69.7#17/26 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Multimodal | Not ranked | 74.4#28/58 | 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 |
| Instruction following | Not ranked | 93.3#9/157 | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | 91.2Unranked · 3 rankable rows | Not comparableProvisional lane · 0 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 4 Carbon has the lower modeled cost
Costs use the listed standard API rates.
50K fresh input + 3K output tokens
Gemini 4 Carbon has the lower modeled cost
Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
Gemini 4 Carbon has the lower modeled cost
Costs use the listed standard API rates.
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 4 Carbon
Not sourced
Google Gemini 4 Carbon launchGPT-5.5
Gemini 4 Carbon
Not sourced
GPT-5.5
gpt-5.5
OpenAI pricingA missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Gemini 4 Carbon
$0.1 per 1M cached input tokens
Google Gemini 4 Carbon launch postGPT-5.5
$0.5 per 1M cached input tokens
OpenAI pricingGemini 4 Carbon
Not sourced
GPT-5.5
Not sourced
Gemini 4 Carbon
Not sourced
GPT-5.5
Not sourced
Gemini 4 Carbon
Limited Access · Fairwind Program
Google Gemini 4 Carbon launchGPT-5.5
Not sourced
Gemini 4 Carbon
Reasoning
GPT-5.5
Reasoning
Gemini 4 Carbon
Proprietary
GPT-5.5
Proprietary
Gemini 4 Carbon
Proprietary
GPT-5.5
Proprietary
Gemini 4 Carbon
2026-10-28
GPT-5.5
2026-04-23
Gemini 4 Carbon has the higher public point estimate, 88.42 versus 68.83. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.
Gemini 4 Carbon has the higher public coding point estimate, 82.2 to 62.4, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Gemini 4 Carbon is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
For the stated presets, chat costs $0.007 on Gemini 4 Carbon and $0.02 on GPT-5.5; repository review costs $0.13 and $0.34; the cache-heavy agent loop costs $0.16 and $0.5. Costs use the listed standard API rates.
A complete documented context-window comparison is not available.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
AutomationBench
Not directly comparable
OSWorld 2.1 Offline
Not directly comparable
CWE-bench v1
Not directly comparable
ExploitGym
Gemini 4 Carbon leads this result
Terminal-Bench 2.0
Not directly comparable
CyberGym
Not directly comparable
BrowseComp
Not directly comparable
OSWorld-Verified
Not directly comparable
MCP Atlas
Not directly comparable
Toolathlon
Not directly comparable
τ²-bench results
Not directly comparable
Gert Labs
Not directly comparable
ResearchClawBench
Not directly comparable
OSWorld 2.0
Not directly comparable
JobBench
Not directly comparable
Terminal-Bench 2.1 (Vals)
Not directly comparable
ApprenticeBench
Not directly comparable
MazeBench
Not directly comparable
FrontierSWE v2
Not directly comparable
SWE-bench Pro
Not directly comparable
Terminal-Bench 2.0
Not directly comparable
Vibe Code Bench
Not directly comparable
React Native Evals
Not directly comparable
cursorBench31
Not directly comparable
CursorBench 3.2
Not directly comparable
FrontierCode 1.1 Main
Not directly comparable
LiveCodeBench (Vals)
Not directly comparable
SWE-bench (Vals)
Not directly comparable
PostTrainBench v1.1
Not directly comparable
FrontierCode 2.0
Not directly comparable
GraphWalks BFS 256K–1M
Not directly comparable
MRCR v2 64K-128K
Not directly comparable
MRCR v2 128K-256K
Not directly comparable
ARC-AGI-2
Not directly comparable
ARC-AGI-3
Not directly comparable
ARC-AGI-1
Not directly comparable
Chartography (no tools)
Not directly comparable
Blueprint-Bench 2
Not directly comparable
ExtractBench (LlamaIndex)
Not directly comparable
ParseBench
Not directly comparable
MMMU-Pro
Not directly comparable
MMMU-Pro w/ Python
Not directly comparable
OfficeQA Pro
Not directly comparable
GPQA
Not directly comparable
GPQA-D
Not directly comparable
HLE
Not directly comparable
HLE w/o tools
Not directly comparable
GPQA Diamond (Vals)
Not directly comparable
MMLU-Pro (Vals)
Not directly comparable
Gray Swan IPI (15 attempts)
Not directly comparable
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 6,100+ subscribers.
Last updated October 28, 2026
