Compare / Gemini 2.5 Flash Preview (09-2025) vs Gemini 3.8 Flash
Gemini 2.5 Flash Preview (09-2025) vs Gemini 3.8 Flash
Which is cheaper, which is more capable, and which does better at each kind of work, from 14 results on 3 sources that measured both models.
Of the 14 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).
Gemini 2.5 Flash Preview (09-2025) or Gemini 3.8 Flash? The short answer
Pick Gemini 3.8 Flash: it scores higher (Intelligence Index 40.9 against 15.5) at about the same price.
- Gemini 3.8 Flash scores higher on the Artificial Analysis Intelligence Index: 40.9 against 15.5.
- Gemini 3.8 Flash is newer: released 2 Sep 2026, against 25 Sep 2025 for Gemini 2.5 Flash Preview (09-2025).
- On third-party benchmarks, on the results both models have: Gemini 3.8 Flash does better for reasoning and knowledge, writing and sticking to the facts.
Head to head
Every fact side by side
| Gemini 2.5 Flash Preview (09-2025) | Gemini 3.8 Flash | |
|---|---|---|
| Price per million tokens | Not on sale | $1.50 in · $7.50 out |
| Blended (3 in : 1 out) | – | $3.00 |
| Cheapest host, blended | – | $1.50 (Google) |
| Intelligence Index | 15.5 | 40.9 |
| Spring Prompt overall | – | 58 |
| Output speed, tokens a second | – | 141 |
| Context window | – | 1,048,576 tokens |
| Released | 25 Sep 2025 | 2 Sep 2026 (newer) |
| Weights | Closed | Closed |
| Developer based in | the US | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
From our research
- Gemini 4 Hydrogen got faster: we reran our speed chess test 16 Nov 2026
- Muse Ember review: GPT-6 Luna's price, better at analysis, and no Gemini Flash against a clock 10 Nov 2026
- Kimi K3.1 Pro first look: the best open-weight model on our board, and very slow 5 Nov 2026
- DeepSeek-V4.1-Flash-1017 review: better at analysis, much worse at sticking to the product data 19 Oct 2026
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Gemini 2.5 Flash Preview (09-2025) | Gemini 3.8 Flash |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 133 · 7 of 387 measured | 1,403 | 1,495high reasoning |
| Overall · Text factualityArena rating, higher is better 105 · 16 of 163 measured | 1,425 | 1,477high reasoning |
| Business, management and finance · TextArena rating, higher is better 135 · 21 of 380 measured | 1,403 | 1,481high reasoning |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 131 · 25 of 249 measured | 15.5thinking reasoning | 40.9high reasoning |
| GPQA Diamond% of questions, higher is better 112 · 2 of 232 measured | 79.3%thinking reasoning | 95.3%high reasoning |
| Humanity's Last Exam% of questions, higher is better 139 · 18 of 250 measured | 13.8%thinking reasoning | 47.8%high reasoning |
| UGI Leaderboard reported by UGI Leaderboard · 3 results | ||
| Writing score · UGIscore out of 100, higher is better 89 · 1 of 230 measured | 44.9no reasoning | 78.6medium reasoning |
| Requested-length error · UGI% off the requested word count, lower is better 111 · 57 of 230 measured | 20.0%no reasoning | 13.0%high reasoning |
| Style adherence · UGIscore from 0 to 1, higher is better 117 · 132 of 230 measured | 0.35thinking reasoning | 0.34high reasoning |
Show all 14 results (5 more)
| Measure | Gemini 2.5 Flash Preview (09-2025) | Gemini 3.8 Flash |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results | ||
| Creative writing · TextArena rating, higher is better 120 · 7 of 385 measured | 1,381 | 1,485high reasoning |
| Expert prompts · TextArena rating, higher is better 123 · 13 of 338 measured | 1,432 | 1,526high reasoning |
| Instruction following · TextArena rating, higher is better 124 · 12 of 387 measured | 1,400 | 1,486high reasoning |
| Writing, literature and language · TextArena rating, higher is better 116 · 8 of 386 measured | 1,394 | 1,487high reasoning |
| Artificial Analysis reported by Artificial Analysis · 1 result | ||
| Long-context reasoning (AA-LCR)% of questions, higher is better 111 · 7 of 241 measured | 71.0%thinking reasoning | 84.0%medium reasoning |
Compare them with other models
Quick answers
Is Gemini 2.5 Flash Preview (09-2025) better than Gemini 3.8 Flash?
Pick Gemini 3.8 Flash: it scores higher (Intelligence Index 40.9 against 15.5) at about the same price.
Which is better for reasoning and knowledge, Gemini 2.5 Flash Preview (09-2025) or Gemini 3.8 Flash?
Gemini 3.8 Flash ranks 15 of 188 models on sale for hard questions across science, maths and general knowledge, against 116 for Gemini 2.5 Flash Preview (09-2025).
Which is better for writing, Gemini 2.5 Flash Preview (09-2025) or Gemini 3.8 Flash?
Gemini 3.8 Flash ranks 3 of 170 models on sale for what people prefer in blind comparisons, and judged writing quality, against 93 for Gemini 2.5 Flash Preview (09-2025).
Which is better for sticking to the facts, Gemini 2.5 Flash Preview (09-2025) or Gemini 3.8 Flash?
Gemini 3.8 Flash ranks 12 of 140 models on sale for summarising without inventing things, and factual answers, against 109 for Gemini 2.5 Flash Preview (09-2025).
Run Gemini 2.5 Flash Preview (09-2025) and Gemini 3.8 Flash on your own prompt.
Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.