Gemini 4 Carbon
Newgoogle/gemini-4-carbon-20261110
Google Gemini 4 Carbon is a multimodal model that accepts audio, file, image, text, and video inputs and returns text outputs. It has a context length of 1,048,576 tokens, supports tools and reasoning, and is not an open-weight model. Pricing is set at $2 per million input tokens and $10 per million output tokens.
- tool use
- structured outputs
- reasoning
At a glance
Specifications
What the API gives you
- Context window
- 1.0M tokens
- Max output
- 944K tokens
- Input price
- $2 / 1M tokens
- Output price
- $10 / 1M tokens
- Cache read
- $0.100 / 1M tokens
- Modalities
- audio, file, image, text, video → text
- Knowledge cutoff
- —
- Released
- Nov 10, 2026
Measured evidence
Public benchmark record
- Design Arena Coding
- 1348 Elo2026-11-15
- Design Arena UI
- 1347 Elo2026-11-15
- Design Arena Website
- 1319 Elo2026-11-15
- FACTS Grounding
- 77.2%2026-10-29evaluated 2026-10-29
- FrontierMath Tiers 1-3 (Epoch AI)‡
- 90.5% ± 1.72026-11-02high effort, evaluated 2026-11-02
- GPQA Diamond (Epoch AI)‡
- 96.5% ± 1.22026-11-02high effort, evaluated 2026-11-02
- LMArena Agent‡
- 172026-11-11high effort
- LMArena Agent Steerability‡
- 19.12026-11-11high effort
- LMArena Agent Task Outcome‡
- 18.42026-11-11high effort
- LMArena Agent Tool Hallucination‡
- 0.52026-11-11high effort
- LMArena Creative Writing‡
- 1533 Elo2026-11-11high effort
- LMArena Instruction Following‡
- 1539 Elo2026-11-11high effort
- LMArena Long Query‡
- 1549 Elo2026-11-11high effort
- LMArena Text‡
- 1526 Elo2026-11-11high effort
- LMArena Text Factuality‡
- 1521 Elo2026-11-11high effort
- LMArena WebDev‡
- 1768 Elo2026-11-11high effort
- OTIS Mock AIME (Epoch AI)‡
- 100% ± 02026-11-02high effort, evaluated 2026-11-02
- SimpleQA Verified (Epoch AI)‡
- 77.7% ± 1.32026-11-02high effort, evaluated 2026-11-02
- Terminal-Bench (4.1)‡
- 64.5%2026-11-15Antigravity CLI, high effort, evaluated 2026-11-14
LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference. ‡ Submitted scaffold plus model result. Compare the harness and effort settings, not only the model name.
Task evidence
How broad is the support?
Each status is exact to this model version and the selected task. Official release claims remain provisional.
- Best LLM for Frontend & UIEstablished evidence
82% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for Grounded RAGEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 7 configured evaluation feeds currently have results.
- Best LLM for a General AssistantEstablished evidence
59% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for Autonomous AgentsEstablished evidence
54% of intended weighted task evidence across 2 independent sources. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for Hermes AgentEstablished evidence
48% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for OpenHandsEstablished evidence
48% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
Show all 39 tasks
- Best LLM for GooseEstablished evidence
48% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for OpenClawEstablished evidence
48% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for CodingEstablished evidence
47% of intended weighted task evidence across 3 independent sources. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for Document SummarizationEstablished evidence
45% of intended weighted task evidence across 2 independent sources. 5 of 6 configured evaluation feeds currently have results.
- Best LLM for SummarizationEstablished evidence
45% of intended weighted task evidence across 2 independent sources. 5 of 6 configured evaluation feeds currently have results.
- Best LLM for RAG & Long ContextEstablished evidence
44% of intended weighted task evidence across 2 independent sources. 5 of 6 configured evaluation feeds currently have results.
- Best LLM for AI AgentsEstablished evidence
41% of intended weighted task evidence across 2 independent sources. 9 of 9 configured evaluation feeds currently have results.
- Best LLM for Logic & Problem SolvingEstablished evidence
41% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Competition MathEstablished evidence
37% of intended weighted task evidence across 3 independent sources. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for Agentic CodingEstablished evidence
35% of intended weighted task evidence across 2 independent sources. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for Math & ReasoningEstablished evidence
32% of intended weighted task evidence across 4 independent sources. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for Business WritingProvisional evidence
67% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Marketing CopyProvisional evidence
67% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Chat & RoleplayProvisional evidence
67% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Roleplay & CharacterProvisional evidence
67% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Creative WritingProvisional evidence
56% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for WritingProvisional evidence
56% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Tool & Function CallingProvisional evidence
38% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Search. 6 of 7 configured evaluation feeds currently have results.
- Best LLM for Code CompletionProvisional evidence
36% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 3 of 3 configured evaluation feeds currently have results.
- Best Open-Weight LLM for Self-HostingProvisional evidence
27% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 2 of 3 configured evaluation feeds currently have results.
- Best Value LLMProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 2 of 3 configured evaluation feeds currently have results.
- Best Local LLM for CodingProvisional evidence
21% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Massive ContextProvisional evidence
20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or LongBench v2. 2 of 3 configured evaluation feeds currently have results.
- Best LLM for Debugging & Code ReviewProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Fast SummariesProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 2 of 3 configured evaluation feeds currently have results.
- Best LLM for Literary TranslationProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Everyday TranslationProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for TranslationProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Computer & Browser UseProvisional evidence
17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or OSWorld Verified. 3 of 3 configured evaluation feeds currently have results.
- Best LLM for JSON & Schema OutputProvisional evidence
12% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Structured Data ExtractionProvisional evidence
12% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 3 of 4 configured evaluation feeds currently have results.
- Best Cheap LLM for High VolumeProvisional evidence
10% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 2 of 3 configured evaluation feeds currently have results.
- Cheapest Capable LLMProvisional evidence
10% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 2 of 3 configured evaluation feeds currently have results.
Official release evidence · provisional
What the launch material actually says
Verbatim claims from the model provider. These help on launch day but do not count as independent confirmation.
“Gemini 4 Carbon, our strongest model with frontier performance in cybersecurity and long-horizon capabilities.”
Google · Gemini 4 Carbon: New model for developer agents“Built for sustained agentic workflows, Carbon tackles complex, multi-step problems across software engineering, full-stack web development, scientific research, and shows strong cybersecurity capabilities.”
Google · Gemini 4 Carbon: New model for developer agents
Recommendation coverage
Where this model appears
- #1Best LLM for Document Summarization
- #1Best LLM for Grounded RAG
- #1Best LLM for Logic & Problem Solving
- #1Best LLM for a General Assistant
- #1Best LLM for Summarization
- #1Best LLM for Massive Context
- #1Best LLM for RAG & Long Context
- #2Best LLM for Competition Math
- #2Best LLM for Math & Reasoning
- #3Best LLM for Hermes Agent
- #3Best LLM for AI Agents
- #3Best LLM for Autonomous Agents
- #3Best LLM for Computer & Browser Use
- #3Best LLM for OpenClaw
- #3Best LLM for Goose
- #3Best LLM for OpenHands
- #4Best LLM for Frontend & UI
- #4Best LLM for Agentic Coding
- #4Best LLM for Coding
- #7Best LLM for Tool & Function Calling
Compare head to head
Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.