Get started

Gemini 4 Carbon

Sold by 5 sellers, timed at 41.7 tok/s, scored 60.0 on the Intelligence Index, on 10 Design Arena boards, and with results on 1 openly licensed board.

retrieved sha256 · 0bb53a77713f

Models/Google/Gemini 4 CarbonUSD per 1M tokens · as OpenRouter routes it
google/gemini-4-carbonGoogle · base listing · Gemini tokenizer
OpenRouter list price · the routed rate, as the catalog carried itretrieved 17 Nov 16:29 UTC from openrouter.ai ↗
Input$2.00per 1M tokens
Output$10.00per 1M tokens
Cached input$0.10per 1M tokens
Context1.05M1,048,576 tokens
The other meters · every further rate the list publishes for this id, in the unit it states4 meters
Cache write$0.1875per 1M tokens
Reasoning tokens$10.00per 1M tokens
Audio input$2.00per 1M tokens
Web search$0.014per search
What the list says · the fields OpenRouter publishes for this id10 fields
AuthorGoogle
Listed
Canonical idgoogle/gemini-4-carbon-20261110
Context1,048,576 tokens
Max output943,718 tokens
Modalitiesaudio, file, image, text, video → text
TokenizerGemini
Reasoningalways on · effort high, medium, low, medium by default
Moderationnot moderated by OpenRouter
Other listingbatch variant

What sellers charge

5 sellers · 3.6x from cheapest to dearest

One tick per seller, on a log scale of input price per million from $1.00 at Google AI Studio to $3.60 at Google AI Studio. 1 seller undercuts the $2.00 listed; 2 are dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.

Measured speed

106,752 requests · 7 endpoints · 30-minute window
Throughput41.7 tok/sMedian output rate, weighted across the 7 endpoints serving it by the requests each answered.the speed board
First token8.86sMedian wait before the first token, over the same window and weighted the same way.how Vitna reads a figure
Endpoint spread33.5 to 72The slowest and the fastest route serving this model, in tokens per second. Speed is a property of the route, not of the weights.the board, fastest first

OpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.

Every endpoint

5 priced · 7 timed · cheapest first
  1. Google AI Studiogoogle-ai-studio/flex · 50% off$1.00$5.00$0.05not declared944knot matchednonenonenot matched100%
  2. Googlegoogle-vertex/global · listed · 50% off$2.00$10.00$0.10not declared944knot matchednonenonenot matched99.79%
  3. Google AI Studiogoogle-ai-studio · listed · 50% off$2.00$10.00$0.10not declared33knot matchednonenonenot matched100%
  4. Googlegoogle-vertex/global/priority · 50% off$3.60$18.00$0.18not declared944knot matchednonenonenot matched98.71%
  5. Google AI Studiogoogle-ai-studio/priority · 50% off$3.60$18.00$0.18not declared944knot matchednonenonenot matched100%
  6. Googlenot matchednot matchednot matchednot matchedno offer33.5 tok/s128.35s156 reqno offer
  7. Googlenot matchednot matchednot matchednot matchedno offer42 tok/s149.1s98,271 reqno offer
  8. Googlenot matchednot matchednot matchednot matchedno offer72 tok/s194.42s30 reqno offer
  9. Googlenot matchednot matchednot matchednot matchedno offer36 tok/s93.11s5 reqno offer
  10. Google AI Studionot matchednot matchednot matchednot matchedno offer44 tok/s236.05s1,296 reqno offer
  11. Google AI Studionot matchednot matchednot matchednot matchedno offer36 tok/s87.53s3,267 reqno offer
  12. Google AI Studionot matchednot matchednot matchednot matchedno offer39 tok/s104.97s3,727 reqno offer

Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 4 of the 7 timed routes are slower than it, and the middle route runs at 39 tok/s. The widest median-to-p90 gap is Google AI Studio's, 4.5x.

Google publishes 2 offers and had 4 routes timed; Google AI Studio publishes 3 offers and had 3 routes timed. The two documents share no route id, so those offers and routes are printed apart rather than paired.

Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 7 of the 7 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: Google AI Studio (google-ai-studio/priority) at 53.33% on 18 Nov 2026, Google (google-vertex/global) at 54.65% on 18 Nov 2026, Google (google-vertex/global/priority) at 74.52% on 19 Nov 2026, Google AI Studio (google-ai-studio) at 83.08% on 11 Nov 2026, and 3 more. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.

Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .

What each seller accepts

12 parameters · 5 sellers
  1. include_reasoningevery seller
  2. max_tokensevery seller
  3. reasoningevery seller
  4. reasoning_effortevery seller
  5. response_formatevery seller
  6. seedevery seller
  7. structured_outputsevery seller
  8. temperatureevery seller
  9. tool_choiceevery seller
  10. toolsevery seller
  11. top_pevery seller
  12. stop2 of 5not Google AI Studio (google-ai-studio), Google AI Studio (google-ai-studio/flex), and Google AI Studio (google-ai-studio/priority)

11 of these 12 are accepted by every one of the 5 sellers that publish a list; the other 1 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.

Price position

342 priced models

Blended $4.00 per million tokens, 3 in to 1 out: cheaper than 55 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.

Measured quality

145 models with an index · 132 on a design arena board

Each mark is one board this model stands on, placed at the share of that board it scores above, and led to its number in the list under the rule. The columns and the ledger below carry the figures.

Intelligence Index60.0Position 2 of 58, above 97% of the models scored.the intelligence leaderboard
Coding Indexnot scoredArtificial Analysis has not scored this model on its Coding Index.the coding leaderboard
Agentic Indexnot scoredArtificial Analysis has not scored this model on its Agentic Index.the agentic leaderboard
  1. codetovideo · Agents1287 Elowin rate 62.6% · source rank 5the board, 26 ranked
  2. Mobile apps · Agents1292 Elowin rate 62.6% · source rank 4the board, 49 ranked
  3. pdf · Agents1305 Elowin rate 65.7% · source rank 3the board, 36 ranked
  4. spreadsheet · Agents1220 Elowin rate 48.3% · source rank 13the board, 32 ranked
  5. 3D scenes · Models1424 Elowin rate 68.7% · source rank 6the board, 113 ranked
  6. Code · Models1341 Elowin rate 59.0% · source rank 9the board, 120 ranked
  7. Game development · Models1332 Elowin rate 57.5% · source rank 12the board, 120 ranked
  8. SVG · Models1400 Elowin rate 69.3% · source rank 1the board, 83 ranked
  9. UI components · Models1345 Elowin rate 59.5% · source rank 8the board, 116 ranked
  10. Websites · Models1311 Elowin rate 55.5% · source rank 10the board, 126 ranked
Epoch Capabilities Index · a fit by a third party over every benchmark this model has results onthe whole index, with intervals
Index169.7166.6 to 173.1
Position3 of 258above 99% of the models placed
Results behind it17benchmark results
As Epoch names itGemini 4 Carbonreleased 10 Nov 2026

Position 3 of the 258 models Epoch has placed, which is not the population of either board above, on a scale anchored at Claude 3.5 Sonnet 130 and GPT-5 150. Epoch states that absolute values are meaningless by themselves and that comparisons between models are what the index is for, that values are not linearly related to benchmark accuracy, and that a highly specialised model may score low while being capable in its own domain.

Benchmark results

1 of 6 boards · 10 results
Epoch AI Benchmarking Hub · publisher-run, CC-BY-4.010 of 22 tasks · at high effort

Mean accuracy, 0 to 1, on the task the row names, as Epoch AI runs it and averaged over repeated runs. All eighteen of their tasks, including the held-out FrontierMath tiers, named as they name them. The model id carries the effort setting the run used. Epoch AI Benchmarking Hub names this run gemini-4-carbon_high. All runs is how many runs the publisher scored on the task, every model's. Read from epoch.ai ↗, which dates its own data . Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.

1 of the 6 openly licensed boards Vitna mirrors carries a result for Gemini 4 Carbon, each printed as its publisher published it and none added to another. MTEB, LiveBench, HELM, Vectara hallucination leaderboard and Aider polyglot have not published one under a name that matches its listing, and names are matched to the catalog exactly and never guessed. A task that is a link opens its column on /capability, with every run scored on it; an effort that is a link opens its run's row there, where the run is ranked on Vitna's scale. Every board, with its licence and what its publisher measures, is on /benchmarks.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Source: the Epoch AI Capabilities Index (epoch.ai/eci), as of 20 Nov 2026. 258 models placed over 2917 benchmark results.
Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
Vitna ran none of these evaluations and adjusted none of these figures.