Skip to content
fedi.software

Gemini 1.5 Flash-8B (001) in fedi-index

The score in every slice, what it is made of, value, local run and history.

fedi-index · General

D 1,217 ±4

Percentile 25 · #224

People prefer / Solves tasks

1,221 / 1,196

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,217 [1,213–1,220] Percentile 25 · #224 · components: 7
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,226 ±2.18 1,226 ±2 22.6% 2026-11-11 —
LMArena text/hard_prompts 1,209 ±3.39 1,223 ±3 22.2% 2026-11-11 —
Epoch AI gpqa_diamond decay ×0.25 0.33 ±0.02 1,193 ±11 13.0% 2025-03-12 —
LMArena text/instruction_following 1,199 ±3.03 1,222 ±3 11.1% 2026-11-11 —
LMArena text/expert 1,185 ±6.65 1,219 ±5 10.2% 2026-11-11 —
LMArena text/multi_turn 1,185 ±4.29 1,204 ±4 5.4% 2026-11-11 —
LMArena text/longer_query 1,218 ±4.30 1,231 ±4 5.4% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Coding 1,217 [1,210–1,223] Percentile 22 · #223 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,218 ±4.10 1,219 ±4 88.3% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Math 1,207 [1,200–1,214] Percentile 24 · #195 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,207 ±4.35 1,208 ±4 88.1% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Vision 1,241 [1,232–1,250] Percentile 16 · #94 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,044 ±5.44 1,244 ±5 93.5% 2026-11-09 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Writing 1,236 [1,230–1,242] Percentile 28 · #190 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,217 ±4.51 1,242 ±5 62.0% 2026-11-11 —
LMArena text/longer_query 1,218 ±4.30 1,231 ±4 31.6% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: 繁體中文 1,237 [1,229–1,246] Percentile 21 · #173 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,231 ±5.94 1,240 ±5 93.6% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: Deutsch 1,232 [1,215–1,249] Percentile 15 · #123 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,206 ±11 1,235 ±9 92.0% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

C Language: English 1,209 [1,204–1,215] Percentile 27 · #197 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,231 ±2.82 1,210 ±3 93.9% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: Español 1,222 [1,196–1,248] Percentile 7 · #140 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,212 ±16 1,225 ±15 89.0% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: Français 1,231 [1,204–1,258] Percentile 9 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,234 ±17 1,234 ±16 88.6% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: 日本語 1,237 [1,222–1,253] Percentile 8 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,150 ±12 1,240 ±9 92.3% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: 한국어 1,235 [1,215–1,255] Percentile 8 · #126 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,140 ±15 1,238 ±11 91.1% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

D Language: Русский 1,256 [1,248–1,263] Percentile 20 · #184 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,236 ±4.45 1,259 ±4 93.7% 2026-11-11 —

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

History

fedi-index by month
The same in a table
Model 2026-11
Gemini 1.5 Flash-8B (001) 1,217 D

Badge

Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.

fedi-index: Gemini 1.5 Flash-8B (001)

Data CC BY 4.0: keep the link to fedi.software.