Skip to content
fedi.software

yi-1.5-34b-chat in fedi-index

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

fedi-index · General

D 1,178 ±3

Percentile 21 · #231 · provisional

People prefer / Solves tasks

1,176 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,178 [1,174–1,181] Percentile 21 · #231 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,173 ±2.55 1,173 ±3 26.6% 2026-10-02 —
LMArena text/hard_prompts 1,160 ±4.22 1,178 ±4 25.7% 2026-10-02 —
LMArena text/instruction_following 1,139 ±3.90 1,166 ±4 12.9% 2026-10-02 —
LMArena text/expert 1,144 ±9.40 1,185 ±8 10.8% 2026-10-02 —
LMArena text/multi_turn 1,153 ±5.74 1,175 ±5 6.1% 2026-10-02 —
LMArena text/longer_query 1,144 ±6.79 1,161 ±6 5.8% 2026-10-02 —

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 1,178 [1,170–1,187] Percentile 20 · #224 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,169 ±5.31 1,176 ±5 89.7% 2026-10-02 —

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 1,186 [1,177–1,196] Percentile 22 · #199 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,182 ±5.64 1,184 ±5 87.6% 2026-10-02 —

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 Writing 1,162 [1,154–1,170] Percentile 20 · #214 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,135 ±5.64 1,158 ±6 62.7% 2026-10-02 —
LMArena text/longer_query 1,144 ±6.79 1,161 ±6 30.5% 2026-10-02 —

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 1,224 [1,214–1,233] Percentile 19 · #180 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,213 ±6.55 1,225 ±5 93.5% 2026-10-02 —

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 1,157 [1,140–1,173] Percentile 11 · #128 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,111 ±11 1,153 ±9 92.1% 2026-10-02 —

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: English 1,193 [1,186–1,200] Percentile 24 · #207 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,215 ±3.47 1,192 ±4 93.8% 2026-10-02 —

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 1,147 [1,121–1,172] Percentile 5 · #141 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,120 ±16 1,140 ±15 89.1% 2026-10-02 —

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 1,167 [1,141–1,193] Percentile 8 · #125 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,155 ±16 1,162 ±15 88.8% 2026-10-02 —

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 1,149 [1,130–1,168] Percentile 6 · #121 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,021 ±14 1,144 ±11 91.4% 2026-10-02 —

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 1,141 [1,123–1,158] Percentile 5 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,005 ±13 1,135 ±10 91.9% 2026-10-02 —

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 1,134 [1,124–1,144] Percentile 8 · #216 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,091 ±6.06 1,129 ±5 93.4% 2026-10-02 —

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 2026-10
yi-1.5-34b-chat 1,178 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: yi-1.5-34b-chat

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