gemini-3.8-flash vs grok-4.3
gemini-3.8-flash vs grok-4.3: $0.75/$3.75 vs $1.25/$2.5 per 1M, the same workload priced on both; gemini-3.8-flash is 19% cheaper per request.
| gemini-3.8-flash | grok-4.3 | |
|---|---|---|
| Provider · tier | gemini · fast | xai · mid |
| Input | $0.75 | $1.25 |
| Cached input | $0.075 | $0.2 |
| Cache write | none | none |
| Output | $3.75 | $2.5 |
| Long-context tier | none | > 200,000: $2.5 / $5 |
| Per request (2,000 in / 300 out) | $0.0026249999999999997 | $0.0032500000000000003 |
| Per month at 100k requests | $263 | $325 |
| …with the static prefix cached | $161 | $168 |
| Cheaper sibling (cascade) | gemini-3.5-flash-lite | — |
Through TokenZero
Point your SDK at the gateway and keep both keys; the cache formatter, response cache and advisory routing act on every request, and each one comes back with a signed receipt showing the list-price baseline, actual cost and the saving.
OPENAI_BASE_URL=https://tokenzero.sh/w/<key>/v1 ANTHROPIC_BASE_URL=https://tokenzero.sh/w/<key>
Questions
Which is cheaper, gemini-3.8-flash or grok-4.3?
At list prices (table 2026-10-26) gemini-3.8-flash is 19% cheaper per request on a 2,000-in / 300-out workload. Quality is a separate question: TokenZero's cascade only routes to a sibling it scores as sufficient and measures parity with shadow evaluation.
Where do these prices come from?
The providers' own pricing pages, re-verified on 2026-10-26; the gateway signs receipts with the same table.