Google · Chat
Gemini 3 Flash
Newer version: Gemini 3.8 Flash. We compared release dates and version names in our catalogue.
Google most intelligent model built for speed, combining frontier intelligence with superior search and grounding.
- llminfo IQ
- 132.0
- Maker's price
- $1.125 blended / 1M
- Max context
- 1M
- Listings
- 67
Evidence and sources
Advantages and trade-offs
Compared with 10 chat models, picked mainly by popularity and IQ
Advantages
Advertised computer use support
Computer use advertised by 2 of the 9 peers that list their features.
Trade-offs
No open weights
The weights aren't public, so you can't run it yourself.
Input & output
as stated by the model's providersCapabilities
Supported operations and provider-stated features
Listings
| 1M | $0.50>200k $0.50cache read $0.05batch $0.25 | $3.00batch $1.50 | $1.00 | $0.10 | — | |
| 1M | $0.25cache read $0.025 | $1.50 | — | — | $0.00000025 | |
| 1M | $0.25 | $1.50 | $0.50 | — | $0.00000025 | |
| 1M | $0.25 | $1.50 | $0.50 | — | $0.00000025 | |
| 1M | $0.25cache read $0.025 | $1.50cache write $0.50 | — | — | — | |
| 1M | $0.25cache read $0.05 | $1.50 | — | — | — | |
| 1M | $0.40 | $2.40 | — | — | — | |
| —Not published by the provider | $0.40 | $2.40 | — | — | — | |
| 1M | $0.404cache read $0.0404 | $2.4242 | — | — | — | |
| 1M | $0.475cache read $0.0475 | $2.85 | — | — | — |
Showing 10 of 57: the maker's own listing first, then the cheapest per 1M input tokens from each provider. One listing with a suspiciously low price is in the full list only.
Also sold by 302ai, Abacus, Aihubmix, Atlascloud, Cloudflare AI Gateway, Cometapi, Crossmodel, Databricks, Eden AI, Fastrouter, Flowdot, Frogbot, Gmicloud, Llmgateway, Llmgateway Providers, Merge Gateway, NanoGPT, Neon, Neosantara, Ofox, Opper, Orcarouter, Perplexity, Perplexity Agent, Prime, Qihang AI, Replicate, T Systems, Tokenmix, Venice, Wavespeed, Writer, Zenmux.
10 listings without a price are hidden.
Benchmarks
Named effort levels run from lowest to highest. Select a result to see its source details.
Human preference
Show 117 more resultsHide 117 results
Agentic
Show results from 2 providers and OpenRouter's automatic routingHide results from 2 providers and OpenRouter's automatic routing
- OpenRouter's automatic routing
- Google AI Studio
Show 7 more resultsHide 7 results
| Benchmark | Score |
|---|---|
| Tax Agent Bench | |
| Agent Poker Bench | |
| Finance Agent (v2) | |
| Harvey's Legal Agent Benchmark v1 |
Show 2 more benchmarksHide 2 benchmarks
| Benchmark | Score |
|---|---|
| Terminal-Bench 2.0 | |
| Terminal-Bench 2.1 |
Coding
| Benchmark | Minimal | not stated |
|---|---|---|
| Coding | ||
| Text Factuality - Coding | ||
| Webdev | ||
| Webdev - Image to Webdev |
Show 3 more benchmarksHide 3 benchmarks
| Benchmark | Minimal | not stated |
|---|---|---|
| Webdev - Webdev | ||
| Webdev - Webdev Html | ||
| Webdev - Webdev React |
Show 10 more resultsHide 10 results
Multilingual
Show 34 more resultsHide 34 results
Long context
Vision
| Benchmark | w. sub @ 1fps | wo sub @ 1fps |
|---|---|---|
| Video-MME-v2 | ||
| Avg Acc | ||
| Capability Consistency | ||
| Level 1 |
Show 3 more benchmarksHide 3 benchmarks
| Benchmark | w. sub @ 1fps | wo sub @ 1fps |
|---|---|---|
| Level 2 | ||
| Level 3 | ||
| Reasoning Coherence |
Reasoning
OCR
Show 3 more resultsHide 3 results
Math
General
Factuality
Science
Show results from 2 providers and OpenRouter's automatic routingHide results from 2 providers and OpenRouter's automatic routing
- Google AI Studio
- OpenRouter's automatic routing
Instruction following
Knowledge
Other
| Benchmark | Score |
|---|---|
| arak (bleu, 0-shot) | |
| conversation (bleu, 0-shot) | |
| dream (exact_match_mcqa, 0-shot) | |
| email (bleu, 0-shot) |
Show 18 more benchmarksHide 18 benchmarks
| Benchmark | gemini-3-flash-preview |
|---|---|
| 4: ALT_환각방지 | |
| 4 subset: haerae_bench_v1_rc | |
| 4 subset: haerae_bench_v1_wo_rc | |
| 4 subset: hrm8k |
Show 32 more benchmarksHide 32 benchmarks
| Benchmark | google/gemini-3-flash-preview |
|---|---|
| ALT_バイアス | |
| ALT_倫理・道徳 | |
| ALT_制御性 | |
| ALT_堅牢性 |
Show 24 more benchmarksHide 24 benchmarks
Show 5 more resultsHide 5 results
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · ArmBench-LLM by Metric-AI. Public numeric primary task results; no separate results licence asserted. · BullshitBench by Peter Gostev, MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · Steel.dev Agent Leaderboards, MIT (licence) · Keelemudelite mõõdupuu: Independent LLM leaderboard for Estonian. Eesti Keele Instituut. https://moodupuu.eki.ee, CC BY 4.0. Latest per-benchmark results selected with partial-run flags retained. (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · HalluHard: Fan, Delsad, Flammarion and Andriushchenko (EPFL), arXiv:2602.01031 · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MLE-bench (OpenAI), MIT (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Outback Yak Research, NepNewsBench v2 research draft. Public numerical aggregates from a candidate preview that the publisher says is not evidence of publication; no release licence assigned and none asserted here. · OmniDocBench (OpenDataLab), Apache-2.0 (licence) · OpenHands Index (All Hands AI), Apache-2.0 (licence) · Shaberi v2.1, Shisa.AI, Apache-2.0. GPT-5.1-judge scores only; judges are not comparable. (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · SWE-bench-Live (Microsoft and contributors), swe-bench-live.github.io. Resolve rates computed from the published report counts. · SWE-fficiency (swefficiency.com) leaderboard materials, CC BY 4.0; Human Expert reference row omitted (licence) · Sierra Research tau2-bench leaderboard submissions (github.com/sierra-research/tau2-bench), MIT (licence) · UzLiB, Tahrirchi (Abror Shopulatov). Public LEADERBOARD.md numbers; repository MIT covers software and documentation, no separate results licence stated and none asserted. · Vectara Hallucination Leaderboard, Apache-2.0 (licence)
Around the web
- HN discussion 50590942Discussion
news.ycombinator.com
- Show HN: AI Spar – Turn conflicting AI answers into a structured debateDiscussion
1 point, 0 comments
news.ycombinator.com
- Show HN: An API for word timestamps from speech, 75% within 20msDiscussion
81 points, 11 comments
news.ycombinator.com
- Discussion
- Testing Gemini 4 Hydrogen and Argon vs. Gen 3 over 1,474 Multi-Turn QueriesDiscussion
3 points, 0 comments
news.ycombinator.com
Also known as: aihubmix/gemini-3-flash-preview, aihubmix/gemini-3-flash-preview-search, databricks-gemini-3-flash, databricks/databricks-gemini-3-flash, gemini 3 flash preview, gemini-3-flash-grounding, gemini-3-flash-minimal, gemini-3-flash-preview, gemini-3-flash-preview-search, gemini-3-flash-preview-thinking, gemini-3-flash-thinking, gemini-3.0-flash-preview, gemini/gemini-3-flash-preview, gmi/google/gemini-3-flash-preview, google-ai-studio/gemini-3-flash-preview, google-gemini-3-flash-preview, google-vertex/gemini-3-flash-preview, google/gemini-3-flash, google/gemini-3-flash-batch, google/gemini-3-flash-preview, google/gemini-3-flash-preview-thinking, google/gemini-3-flash-preview:batch, perplexity/google/gemini-3-flash-preview, replicate/google/gemini-3-flash, vertex/gemini-3-flash-preview, vertex/gemini-3-flash-preview:flex, vertex_ai/gemini-3-flash-preview
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works