GPQA DiamondVerified
95.3%
GPQA Diamond (AA run)
Models / Google
Gemini 3.8 Flash
Gemini 3.8 · Released 2026-09-02
Google's 2 Sep 2026 Flash workhorse. Same intro API price as 3.7 Flash: $0.75 / $3.75 per 1M tokens through 31 Dec 2026. 1M context, 64K output.
Current Google Flash model for coding agents and high-throughput tool use. Prefer sourced AA ledger rows over launch DeepSWE / HLE-Verified / Vals Finance claims.
Composite73
Context1M
Input / 1M$0.75
Output / 1M$3.8
Knowledge cutoff2026-03-01
Output speed132 t/s
Statusflagship
Family profile
Best published score in each covered benchmark family.
80/100 avg
- Reasoning
- 95
- Coding
- 57
- Agentic
- 88
- Long context
- 81
4 tested benchmark families
Editor's note
Google-reported DeepSWE, HLE-Verified 54.9%, Vals Finance Agent V2, and Harvey Legal figures stay vendor-claimed. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where Gemini 3.8 Flash places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | #2/ 184 | 95 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench | Agentic | #10/ 167 | 88 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #15/ 186 | 48 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #15/ 97 | 57 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #29/ 184 | 81 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #35/ 195 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Gemini 3.8 Flash shows up, it's highlighted.
ReasoningGemini 3.8 Flash · 95
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingGemini 3.8 Flash · 57
- 174
- 273
- 370
AgenticGemini 3.8 Flash · 88
- 191
- 289
- 389
Long contextGemini 3.8 Flash · 81
- 189
- 286
- 385
Newest receipts
6 recentLast 120 days
Benchmark rows added to the public ledger for Gemini 3.8 Flash in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 132 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 81.33% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 95.3% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 47.8% | Artificial Analysis | 24 |
| SciCode | Coding | 56.6% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 87.64% | Artificial Analysis | 24 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Gemini 3.8 Flash sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Gemini 3.8 Flash. Scores link to the original source; gaps mean no public row exists yet.
6 of 13 catalog benchmarks have a sourced row for Gemini 3.8 Flash.
Reasoning
Humanity's Last ExamVerified
47.8%
HLE (AA run)
Coding
SciCodeVerified
56.6%
SciCode (AA run)
Agentic
Terminal-BenchVerified
87.64%
Terminal-Bench v2.1 (AA run)
Long context
AA-LCRVerified
81.33%
AA-LCR (AA run)
Performance
AA output speedVerified
132 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Re-checked Google launch post: 3.8 Flash intro API list still $0.75 / $3.75 through 31 Dec 2026. Gemini 3.8 Live family shipped separately on 15 Sep.
- Gemini 3.8 Flash generally available.
Sources
Where the numbers come from
- Google Gemini 3.8 Flash announcement· checked 2026-09-18
- DeepMind Gemini 3.8 Flash model card· checked 2026-09-18
- Artificial Analysis Gemini 3.8 Flash· checked 2026-09-10
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Gemini 3.8 Flash?
Gemini 3.8 Flash is a Google model: Google's 2 Sep 2026 Flash workhorse. Same intro API price as 3.7 Flash: $0.75 / $3.75 per 1M tokens through 31 Dec 2026. 1M context, 64K output.
What is Gemini 3.8 Flash for?
Current Google Flash model for coding agents and high-throughput tool use. Prefer sourced AA ledger rows over launch DeepSWE / HLE-Verified / Vals Finance claims.
What does Gemini 3.8 Flash cost?
Gemini 3.8 Flash lists $0.75 input and $3.8 output per 1M tokens. Pricing last checked 2026-09-18.
How large is the Gemini 3.8 Flash context window?
Gemini 3.8 Flash publishes a 1M token context window.
What is the VerdictPal status of Gemini 3.8 Flash?
Gemini 3.8 Flash carries Flagship status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.