Long Context · Benchmark profile

GDM-MRCR v2 8-needle, 128K average

Google's cumulative eight-needle MRCR v2 score at a 128K average context length, retained as an exact reference track.

Referencev2, 128K averageActive4 models

Data verified 13 Aug 2026 · Methodology 2.2.0

Benchmark score on GDM-MRCR v2 8-needle, 128K average

Leaderboard · 4 models

Gemini 3.7 Flash leads at 97%, followed by GPT-5.6 Terra (93.5%) and Gemini 3.6 Flash (91.8%).

Display onlyLong Contextv2, 128K average

#1Closed

Gemini 3.7 Flash

Google

gemini-3.7-flash

97%

Overall #6 · Context 1.05M tokens

#2Closed

GPT-5.6 Terra

OpenAI

gpt-5.6-terra

93.5%

Overall #9 · Context 1.05M tokens

#3Closed

Gemini 3.6 Flash

Google

gemini-3.6-flash

91.8%

Overall #20 · Context 1.05M tokens

ViewModelsColours

GoogleOpenAIAnthropic

Lumina

GDM-MRCR v2 8-needle, 128K average leaderboard

Accessible data for GDM-MRCR v2 8-needle, 128K average leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Gemini 3.7 Flash	#1	gemini-3-7-flash	Google	97	percent
GPT-5.6 Terra	#2	gpt-5-6-terra	OpenAI	93.5	percent
Gemini 3.6 Flash	#3	gemini-3-6-flash	Google	91.8	percent
Claude Sonnet 5	#4	claude-sonnet-5	Anthropic	81.5	percent

Benchmark score table (4 models)

One best compatible score per canonical product · higher is better

GDM-MRCR v2 8-needle, 128K average leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Gemini 3.7 Flash gemini-3.7-flash	Google	closed	Reference only	97%
#2	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	Reference only	93.5%
#3	Gemini 3.6 Flash gemini-3.6-flash	Google	closed	Reference only	91.8%
#4	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	Reference only	81.5%

The top of this snapshot is led by Gemini 3.7 Flash at 97%; third place is 5.2 points behind. The top-4 spread is 15.5 points.

Configuration-preserving evidencePublished source rows (4)Exact source labels, configurations, versions and dated observations. Same-snapshot canonical/configuration projections are shown once.

Every distinct dated source-published row for GDM-MRCR v2 8-needle, 128K average, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Gemini 3.7 Flash gemini-3-7-flash-medium	Gemini 3.7 Flash	97%	v2, eight-needle, 128K averageGDM-MRCR v2 cumulative eight-needle 128K average	Reference only provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
GPT-5.6 Terra gpt-5-6-terra-max	GPT-5.6 Terra	93.5%	v2, eight-needle, 128K averageGDM-MRCR v2 cumulative eight-needle 128K average	Reference only provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Gemini 3.6 Flash gemini-3-6-flash-high	Gemini 3.6 Flash	91.8%	v2, eight-needle, 128K averageGDM-MRCR v2 cumulative eight-needle 128K average	Reference only provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Claude Sonnet 5 claude-sonnet-5-max	Claude Sonnet 5	81.5%	v2, eight-needle, 128K averageGDM-MRCR v2 cumulative eight-needle 128K average	Reference only provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13

Compare top models

Gemini 3.7 Flash vs GPT-5.6 Terra GPT-5.6 Terra vs Gemini 3.6 Flash Gemini 3.6 Flash vs Claude Sonnet 5

About GDM-MRCR v2 8-needle, 128K average

Definition and scoring

Organisation: Google DeepMind
Category: Long Context
Version: v2, 128K average
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

Google's cumulative eight-needle MRCR v2 score at a 128K average context length, retained as an exact reference track. Every genuine source row stays tied to its exact model label, configuration, benchmark version and evaluation system. The summary chart shows one best compatible score per canonical product; Score 2.0 admits only explicitly mapped, frozen protocols and keeps incompatible configurations separate.

Open benchmark source ↗

FAQ

Quick answers

What does GDM-MRCR v2 8-needle, 128K average measure?

Google's cumulative eight-needle MRCR v2 score at a 128K average context length, retained as an exact reference track.

Which model scores highest on GDM-MRCR v2 8-needle, 128K average?

Gemini 3.7 Flash by Google currently leads with 97%.

How many models are evaluated on GDM-MRCR v2 8-needle, 128K average?

4 canonical model products have a qualifying score, backed by 4 distinct dated source rows.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Long Context · Benchmark profile

GDM-MRCR v2 8-needle, 128K average

Google's cumulative eight-needle MRCR v2 score at a 128K average context length, retained as an exact reference track.

Referencev2, 128K averageActive4 models

Data verified 13 Aug 2026 · Methodology 2.2.0

Benchmark score on GDM-MRCR v2 8-needle, 128K average