Pass IndexThe State of AISign in

Gemini 3.8 Flash

Gemini 3.8 Flash is a model by Google for reasoning and writing code and agentic, it stands 2nd of 69 on Deepswe — Epoch AI, and 14 sellers in the catalogue publish a price for it.

text + image + video + audio + file → text · made by Google

$0.75$3.75per Mtok in / out
Google · 9 sellers

Sold by 17 ways

SellerLaneRate
Google Vertex AI cloudbatch$0.38per Mtok in$1.88per Mtok outraw.githubusercontent.com · read 2026-09-16
Google Vertex AI cloudflex$0.38per Mtok in$1.88per Mtok outraw.githubusercontent.com · read 2026-09-16
Google Vertex AI cloudstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cached$3.75per Mtok out, reasoningmodels.dev · read 2026-09-16
Google Vertex AI cloudpriority$1.35per Mtok in$6.75per Mtok out$0.14per Mtok cachedraw.githubusercontent.com · read 2026-09-16
OpenRouter 0 ways$0.38per Mtok in$1.88per Mtok out$0.075per Mtok cached
OpenRouter aggregatornot on the seller's list todaystandard$0.38per Mtok in$1.88per Mtok out$0.075per Mtok cachedopenrouter.ai · read 2026-09-03
GitHub Copilot aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-16
Google aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
Vercel AI Gateway 0 ways$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cached
Vercel AI Gateway aggregatornot on the seller's list todaystandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
google-vertex aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
kilo 0 ways$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cached
kilo aggregatornot on the seller's list todaystandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
llmgateway aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
llmgateway-providers aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
merge-gateway aggregatorstandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
nano-gpt 0 ways$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cached
nano-gpt aggregatornot on the seller's list todaystandard$0.75per Mtok in$3.75per Mtok out$0.075per Mtok cachedmodels.dev · read 2026-09-03
Requesty 0 ways$0.82per Mtok in$4.12per Mtok out$0.083per Mtok cached
Requesty aggregatornot on the seller's list todaystandard$0.82per Mtok in$4.12per Mtok out$0.083per Mtok cachedmodels.dev · read 2026-09-03
Venice AI aggregatorstandard$0.94per Mtok in$4.69per Mtok out$0.094per Mtok cachedmodels.dev · read 2026-09-16
opencode aggregatorstandard$1.5per Mtok in$7.5per Mtok out$0.15per Mtok cachedmodels.dev · read 2026-09-03

Measured 29 standings

PlaceBoardMetricScore
2ndof 69Deepswe — Epoch AIPass@1 (high)0.738
2ndof 313GPQA Diamondmean_score (high)95.391
4thof 222Chess puzzles — Epoch AImean_score (high)61
5thof 80Simpleqa verified — Epoch AImean_score (high)69.7
7thof 63Design Arena · mobileapps (agents)Elo1268
8thof 150Design Arena · uicomponent (models)Elo1339
8thof 127Mystery game puzzles — Epoch AImean_score (high)47
9thof 74Cursorbench — Epoch AIScore (high)69.2
9thof 402LMArena · TextScore (Elo) (high)1493
11thof 162Design Arena · website (models)Elo1315
12thof 290OTIS Mock AIME 2024-2025mean_score (high)98.889
13thof 156Design Arena · codecategories (models)Elo1324
14thof 184Artificial Analysis · Coding IndexIndex76.3
14thof 155Design Arena · gamedev (models)Elo1340
14thof 62Design Arena · webapps (agents)Elo1255
16thof 64Design Arena · fullstack (agents)Elo1251
17thof 121Webdev arena — Epoch AIArena Score (high)1567.33
18thof 147Design Arena · 3d (models)Elo1325
19thof 106Frontiermath tiers 1 3 v2 — Epoch AImean_score (high)68.421
21stof 64Proofbench — Epoch AIAccuracy48
23rdof 128LMArena · WebDevScore (Elo) (high)1568
25thof 46LMArena · AgentNet Improvement (%)4.71
29thof 302Artificial Analysis · Intelligence IndexArtificial Analysis Intelligence Index (high)41
29thof 43Design Arena · python-pptxslides (agents)Elo1176
30thof 168Scicode — Epoch AIScore (medium)54.398
36thof 143Artificial Analysis · Agentic IndexIndex41.1
38thof 152Design Arena · dataviz (models)Elo1260
39thof 62Frontiermath tier 4 v2 — Epoch AImean_score (high)21.951
40thof 177Critpt — Epoch AIAccuracy (high)18.286

About

Gemini 3.8 Flash — a text model from Google, sold by 9 companies from $0.75 in and $3.75 out per million tokens, placed 2nd of 313 on GPQA Diamond.

It takes text, images, video, audio and files and returns text, with a context window of 1,048,576 tokens. It was published in September 2026. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Nine companies sell it. The cheapest is $0.75 in and $3.75 out per million tokens at GitHub Copilot. Beside the standard rate there are batch, flex and faster lanes. It has been measured on 29 boards, and stands best at 2nd of 313 on GPQA Diamond.

Every current figure

Maker
Google
Register
model
Takes
text + image + video + audio + file
Returns
text
Context
1,048,576 tokens
Published
September 2026
Licence
not read
Sellers
9
Price
$0.75 in and $3.75 out per million tokens — GitHub Copilot
Boards
29
Best place
2nd of 313 — GPQA Diamond