GGoogle · Released Nov 10, 2026
Gemini 4 Carbon
Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex reasoning work. It takes text, image, audio, video and file input, supports long decode...
- #1GPQA Diamond
- #1CritPt
- #1OTIS Mock AIME
- #1APEX-Agents
- Released
- Nov 10 2026
- Context window
- 1M 944K out
- Input / 1M tokens
- $2.00 from $1.00
- Output / 1M tokens
- $10.00 from $5.00
Where to run it
17 providers compared
Cheapest: Google AI Studio at $1.00 / $5.00 per 1M input / output tokens — 75% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| Google AI Studio | $1.00 | $5.00 | $0.05 | 1M | 944K | 100%24h | Cheapest |
| Google AI Studio | $2.00 | $10.00 | $0.10 | 1M | 33K | 99.4%24h | +100% |
| Googleglobal | $2.00 | $10.00 | $0.10 | 1M | 944K | 84.9%24h | +100% |
| Kilo Gateway | $2.00 | $10.00 | $0.10 | 1M | 944K | — | +100% |
| NanoGPT | $2.00 | $10.00 | $0.10 | 1M | 33K | — | +100% |
| Merge Gateway | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| OpenCode | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| Prime Inference | $2.00 | $10.00 | — | 1M | 944K | — | +100% |
| GitHub Copilot | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| Vercel AI Gateway | $2.00 | $10.00 | $0.10 | 1M | 1M | — | +100% |
| CrossModel | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| Vertex | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| Cloudflare AI Gateway | $2.00 | $10.00 | $0.10 | 1M | 66K | — | +100% |
| Googleus | $2.20 | $11.00 | $0.11 | 1M | 66K | 98.6%24h | +120% |
| Googleeu | $2.20 | $11.00 | $0.11 | 1M | 66K | 94.4%24h | +120% |
| Cortecs | $2.23 | $11.15 | $0.11 | 1M | 1M | — | +123% |
| Venice AI | $2.50 | $12.50 | $0.13 | 1M | 66K | — | +150% |
| Googleglobal/priority | $3.60 | $18.00 | $0.18 | 1M | 944K | 100%24h | +260% |
| Google AI Studiopriority | $3.60 | $18.00 | $0.18 | 1M | 944K | 100%24h | +260% |
| LLM Gateway | $4.00 | $20.00 | $0.20 | 1M | 1M | — | +300% |
| Eden AI | $4.00 | $20.00 | $0.20 | 1M | 66K | — | +300% |
| DevPass (LLM Gateway) | $4.00 | $20.00 | $0.20 | 1M | 66K | — | +300% |
Value
Price vs. performance
Gemini 4 Carbon Other Google Other labs Value frontier · GPQA Diamond vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
17 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| FrontierSWESoftware engineering · Ultra-long-horizon engineering tasks with a 20-hour budget each | 61.2% | #5of 21 | Claude Fable 5.269.9% |
| CursorBenchSoftware engineering · Ambiguous multi-file tasks drawn from real Cursor sessions | 60.4% | #1of 20 | This model |
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 64.2% | #3of 105 | Claude Fable 5.267.7% |
| APEX-AgentsAgents & computer use · Long professional tasks in banking, consulting and law | 84.2% | #1of 56 | This model |
| Vending-Bench 2Agents & computer use · Run a simulated vending business for a year (final balance) | $16,484 | #1of 63 | This model |
| SimpleQA VerifiedWriting & content · Factual accuracy on 1,000 short questions | 77.7% | #2of 79 | Claude Fable 5.294.0% |
| Blueprint-Bench 2Design & visual · Turn apartment photos into 2D floor plans | 58.6% | #1of 35 | This model |
| Furniture AssemblyDesign & visual · Spot mistakes in photos of IKEA builds | 77.5% | #6of 37 | Claude Fable 5.290.0% |
| GPQA DiamondReasoning & knowledge · PhD-level biology, chemistry and physics questions | 96.5% | #1of 120 | This model |
| ARC-AGI-2Reasoning & knowledge · Few-shot abstract reasoning on grid puzzles | 94.2% | #2of 76 | GPT-6 Astra95.0% |
| ARC-AGI-1Reasoning & knowledge · Pattern induction from input–output examples | 98.5% | #1of 74 | This model |
| SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people | 92.6% | #2of 71 | Claude Fable 5.294.9% |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 36.0% | #1of 116 | This model |
| Chess PuzzlesReasoning & knowledge · Find the single best move in 100 new puzzles | 78.0% | #2of 82 | Claude Fable 5.279.0% |
| FrontierMathMaths · Expert-written problems, undergrad to early research (tiers 1–3) | 90.5% | #6of 85 | Claude Fable 5.295.1% |
| FrontierMath Tier 4Maths · Exceptionally hard research-level problems | 95.1% | #4of 65 | Claude Fable 5.2100.0% |
| OTIS Mock AIMEMaths · Competition-style problems harder than MATH Level 5 | 100.0% | #1of 118 | This model |
Specs
Details
- Model ID
google/gemini-4-carbon- Developer
- Release date
- Nov 10, 2026 ·
- Context window
- 1,048,576 tokens
- Max output
- 943,718 tokens
- API price
- $2.00 / $10.00 per 1M input / output tokens
- Providers
- 17
- Input types
- audio, file, image, text, video
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- No
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.