ModelBrief · Gemini 4 Carbon · Topic Page
Last updated Nov 13, 2026

Thinking Levels and Pitfalls: Medium Is Cheapest and Most Honest; a Low Output Cap Returns Empty

Published Nov 13, 2026 · Last updated Nov 13, 2026

Carbon's thinking levels are not a simple "higher is better": Artificial Analysis data across all three levels shows the default medium level is simultaneously the cheapest and the most honest — and if you set the output cap wrong, the model spends everything on thinking and returns an empty answer that is still billed. This guide synthesizes eesel.ai's review, Inferera's API migration guide and Quantslant's measurements. The core advice up front: keep the default medium; set max_output_tokens to at least 32,768; set HTTP timeouts to 15–20 minutes; use high only with a larger output budget.

The three levels: medium is the sweet spot

Artificial Analysis Intelligence Index v5.0.1, at list prices (checked by eesel.ai, Nov 11):

Thinking levelAA indexCost per taskMade-up-answer rate
Low54$7.1920%
Medium (default)58$6.594.2%
High60$7.7611%

The counterintuitive part is medium being cheaper than low: weaker settings loop more and retry more, burning more tokens (Hacker News commenters spotted the same oddity). The made-up-rate gap matters more: medium fabricates about once in 25 answers, against roughly two in three for Opus 5.5; high climbs back to 11% — thinking longer does not mean being more honest, and AA-Omniscience likewise measured Carbon (high) hallucinating at 11% versus Argon's 4%. Eesel's scoring puts it vividly: 4.5/5 for hard reasoning and research, 2/5 for anything a person waits on — "a back-office model, not the one that answers live".

The empty-answer mechanism: thinking and answer share the cap

Carbon's max output cap (65,536 streaming / 32,768 non-streaming) covers thinking tokens. Google's long-decode guide says it plainly: "A low value, such as 8,192, stops the model before it finishes thinking" and "Text can be empty if the model was still thinking." Quantslant confirmed it twice in practice: with a 6,000-token cap, Carbon spent 5,808 tokens on reasoning and returned an empty answer, still billed; on a regex task at medium effort it spent all 16,000 tokens on reasoning and returned no code. Lao Zhang's blog hit the same thing: Carbon at high effort ran out of output tokens on a long log before writing a word. The status becomes incomplete — treat it as a cap problem: raise the limit, retry, and accept that the first attempt is already on the bill.

API migration: the 400-error checklist

Carbon is a drop-in replacement for Argon at the request-schema level (change the model ID and go), but these parameters break (Inferera migration guide, Nov 12):

SettingResultCorrect approach
thinking_budgetHTTP 400Use thinking_level: low / medium / high
thinking_level=minimalHTTP 400Carbon has no minimal level; use low
candidate_count > 1HTTP 400Keep it at 1
temperature / top_p / top_kIgnoredRemove them
max_output_tokens too lowSilent truncation; empty text still billedAt least 32,768; larger budget for high
Default HTTP timeoutLong-thinking tasks time out15–20 minutes recommended for streaming long decode

Two more default traps: Carbon defaults to medium — teams migrating from Gemini 3.1 Pro Preview (which defaults to high) drop a level without changing code, which can read as a regression; and the Gemini app's model picker defaults to High, unlike the API default.

The time cost: about a minute to the first word

From the AA model page: Carbon (High) takes about 26–27 minutes per Intelligence Index task (median in its tier: 13.0 minutes) and about 67.7 seconds to the first answer token (median 13.45 seconds) — among the slowest in its price tier. Eesel's hands-on experience matches: about a minute before the first word, ~25.6 minutes per benchmark task on high. The Hacker News thread on the GA post drew 145 comments within a day, with speed and access complaints showing up first. The conclusion is simple: Carbon doesn't belong anywhere a person waits in real time; treat it as a background worker — submit the job and go do something else.

Our take: Carbon's correct setup is four rules: start on the default medium; make the output cap generous rather than tight (at least 32,768); set timeouts to 15–20 minutes; reserve high for the hardest tasks with a full output budget. Break any one of them and you pay real money for it — empty answers are billed too.

Reference source: this article is an original rewrite produced by this site; facts and figures were checked against eesel.ai review and Inferera API migration guide (Nov 11–12, 2026)

← Back to the Gemini 4 Carbon topic page