Skip to content
fedi.software

Llama 2 13B Chat in fedi-index

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

fedi-index · General

D 1,086 ±10

Percentile 10 · #267

People prefer / Solves tasks

1,091 / 1,069

Human votes vs benchmarks, Elo-eq

Value

-242

$0.225 per 1M tokens, median of 2 providers

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,086 [1,076–1,096] Percentile 10 · #267 · components: 7
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,084 ±3.39 1,084 ±3 23.4% 2026-11-11 —
LMArena text/hard_prompts 1,051 ±5.20 1,080 ±5 22.4% 2026-11-11 —
Epoch AI eci 96.66 ±4.35 1,017 ±34 13.8% 2023-07-18 —
LMArena text/instruction_following 1,045 ±4.73 1,077 ±4 11.3% 2026-11-11 —
LMArena text/expert 1,030 ±13 1,092 ±10 8.3% 2026-11-11 —
LMArena text/multi_turn 1,050 ±7.06 1,081 ±6 5.2% 2026-11-11 —
LMArena text/longer_query 1,064 ±8.79 1,087 ±8 4.7% 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

D Coding 1,096 [1,086–1,106] Percentile 9 · #261 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,062 ±6.52 1,080 ±6 87.4% 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 Math 1,089 [1,077–1,100] Percentile 9 · #234 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,065 ±6.84 1,072 ±7 87.0% 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 Writing 1,083 [1,073–1,092] Percentile 9 · #241 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,047 ±6.36 1,067 ±6 63.7% 2026-11-11 —
LMArena text/longer_query 1,064 ±8.79 1,087 ±8 29.2% 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 1,068 [1,054–1,082] Percentile 4 · #211 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,001 ±9.59 1,057 ±8 92.7% 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

D Language: Deutsch 1,079 [1,055–1,102] Percentile 2 · #142 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,009 ±15 1,065 ±13 90.0% 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: English 1,099 [1,092–1,107] Percentile 13 · #235 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,120 ±3.93 1,092 ±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

D Language: Русский 1,105 [1,088–1,122] Percentile 5 · #219 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,054 ±11 1,096 ±10 91.9% 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

History

fedi-index by month
1,085 1,090 1,095 1,100 2026-10 2026-11 Llama 2 13B Chat · 2026-10 · 1,097Llama 2 13B Chat · 2026-11 · 1,086 Llama 2 13B Chat
The same in a table
Model 2026-102026-11
Llama 2 13B Chat 1,0971,086 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: Llama 2 13B Chat

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