Gemini 3.6 Flash (High): benchmark scores, price and ranking

Gemini 3.6 Flash (High) from Google ranks #318 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.9), based on 5 benchmark results captured up to 2026-10-13.

Organisation: Google
Type: Fast / low-cost (flash) model
Open weights: No
Released: 2026-07-21 (source: epoch)
Cheapest API price: $0.750 in / $3.75 out per million tokens (via gemini, observed 2026-10-13T20:47Z)
Aggregate rank: #318 of 743 (score 48.9)

Relative to other models, Gemini 3.6 Flash (High) is strongest on LMArena Document Elo (#5 of 18) and weakest on LMArena Code Elo (#24 of 30).

Across the 5 benchmarks where at least ten models are scored, Gemini 3.6 Flash (High) is in the top 10% on 0 and in the top half on 4.

Its cheapest listed input price, $0.750 per million tokens, is lower than that of 40% of the 427 models with a listed price.

Gemini 3.6 Flash (High) benchmark results

Gemini 3.6 Flash (High) benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
LMArena Document Elo	1,456	#5 of 18	LMArena	2026-10-13
ARC-AGI-2	60.4%	#30 of 87	ARC Prize official leaderboard	2026-10-13
LMArena Text Elo	1,483	#11 of 28	LMArena	2026-10-06
LMArena Vision Elo	1,280	#11 of 22	LMArena	2026-10-05
LMArena Code Elo	1,536	#24 of 30	LMArena	2026-10-06

About Gemini 3.6 Flash (High)

Gemini 3.6 Flash (High) is a smaller, lower-cost model from Google. It was released on 2026-07-21, according to Epoch AI. The cheapest API listing we track charges $0.750 per million input tokens and $3.75 per million output tokens, through gemini. We hold 5 benchmark results for it across 3 areas, and none is in the top tenth of its board.

Where Gemini 3.6 Flash (High) stands

On our aggregate score, Gemini 3.6 Flash (High) is #318 of 743 ranked models, around the middle of the table. Its score is 48.9; the scale is centred on 50, so it sits 1.1 points below the typical model. That beats 57% of the ranked field. Among the 29 ranked Google models it is #16. The next one up is Gemini 2.0 Pro exp-02-05 at 49.4. The next one down is gemini-1.5-flash-exp-0827 at 48.7.

Gemini 3.6 Flash (High), benchmark by benchmark

LMArena Document Elo puts Gemini 3.6 Flash (High) at 1,456, #5 of 18. That beats 72% of the models tested, but 57 points short of Claude Fable 5.1 (max) (1,513). The results either side are gemini-3.5-flash-high (1,461) and Kimi K2.6 (1,451). It covers crowd-voted questions about documents. The board median is 1,440, so Gemini 3.6 Flash (High) is 16 points above it. Source: LMArena, captured 2026-10-13.

On ARC-AGI-2, Gemini 3.6 Flash (High) reaches 60.4%. It ranks #30 of 87, ahead of 66% of the field, and 34.6 points behind GPT-6 Astra (max) at 95.0%. The results either side are DeepSeek V4 Pro 0813 (max) (61.2%) and Kimi K3 (max) (60.4%). It covers abstract reasoning, ARC Prize public eval set. The board median is 22.8%, so Gemini 3.6 Flash (High) is 37.6 points above it. Source: ARC Prize official leaderboard, captured 2026-10-13.

Gemini 3.6 Flash (High) manages 1,483 on LMArena Text Elo, good for #11 of 28. gemini-4-argon-high leads at 1,525, 42 points further up, and 61% of models sit below. The results either side are Gemini 3 Pro (1,485) and GPT-6.1 Sol (max) (1,483). It covers head-to-head chat preference (crowd votes). The board median is 1,478, so Gemini 3.6 Flash (High) is 6 points above it. Source: LMArena, captured 2026-10-06.

LMArena Vision Elo puts Gemini 3.6 Flash (High) at 1,280, #11 of 22. That beats 50% of the models tested, but 16 points short of Gemini 3.7 Flash (high) (1,296). The results either side are GPT-6 Astra (max) (1,284) and gpt-5.2-chat-latest-20260210 (1,278). It covers crowd-voted image understanding. The board median is 1,279, so Gemini 3.6 Flash (High) is 1 points above it. Source: LMArena, captured 2026-10-05.

Gemini 3.6 Flash (High) lands below the middle on LMArena Code Elo: 1,536, #24 of 30, with 77% of the field scoring higher. Claude Opus 5.5 (max) leads at 1,814; the closest result above is step-5-preview-high at 1,570. It covers crowd-voted coding answers. The board median is 1,622, so Gemini 3.6 Flash (High) is 86 points below it. Source: LMArena, captured 2026-10-06.

Strengths and weaknesses

General knowledge and chat: solid, ahead of 61% of the field on average across 3 benchmarks. Its best board here is LMArena Document Elo (#5 of 18) and its weakest is LMArena Vision Elo (#11 of 22).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so Gemini 3.6 Flash (High) costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on qwq-plus ($0.800) and Nova Pro ($0.800).

Release history

The Google model released before it was Gemini 3.5 Flash-Lite (High) on 2026-07-21, which scores 48.0. The next one was gemini-4-argon-high on 2026-09-30, at 59.5.

Reading these numbers

The 5 scores on this page come from 2 sources: LMArena (4), ARC Prize official leaderboard (1). 1 of them come from an official leaderboard and 4 from datasets or third-party runs. The leaderboard also lists 4 other reasoning-effort settings of this model, which redirect here. This page shows the highest. With 5 results the aggregate rank can move a lot when one new score arrives.

Gemini 3.6 Flash (High) is also listed at 4 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.

Models ranked nearby

#315 grok-4.3 (high) (48.9)
#316 claude-opus-4-5-20251101-high-32k (48.9)
#317 DeepSeek V4.1 Flash (high) (48.9)
#319 Claude Opus 4 (48.9)
#320 Gemini 2.0 Flash (Experimental) (48.8)
#321 Phi-3-medium-4k-instruct (48.8)