Skip to content
fedi.software

GLM 5.1 in fedi-index

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

fedi-index · General

A 1,432 ±5

Percentile 83 · #51

People prefer / Solves tasks

1,445 / 1,415

Human votes vs benchmarks, Elo-eq

Value

+36

$2.14 per 1M tokens, median of 42 providers · vendor $2.15

On your hardware

Does not fit the reference setups.

Score decomposition

A General 1,432 [1,427–1,437] Percentile 83 · #51 · components: 9
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 150 ±0.90 1,436 ±7 33.9% 2026-04-07 —
LMArena text/overall 1,461 ±1.87 1,462 ±2 15.6% 2026-10-02 —
LMArena text/hard_prompts 1,472 ±2.17 1,459 ±2 15.6% 2026-10-02 —
LMArena text/instruction_following 1,451 ±2.63 1,461 ±2 7.7% 2026-10-02 —
LMArena text/expert 1,476 ±4.14 1,457 ±3 7.5% 2026-10-02 —
LMArena text/longer_query 1,466 ±2.51 1,462 ±2 3.9% 2026-10-02 —
Epoch AI simpleqa_verified decay ×0.61 0.34 ±0.01 1,404 ±5 3.8% 2026-08-27 —
LMArena text/multi_turn 1,472 ±3.51 1,464 ±3 3.8% 2026-10-02 —
Epoch AI gpqa_diamond decay ×0.26 0.90 ±0.02 1,458 ±10 1.3% 2026-08-10 —

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

A Coding 1,434 [1,430–1,437] Percentile 85 · #43 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,508 ±3.44 1,452 ±1 39.6% 2026-10-01 —
LMArena text/coding 1,486 ±2.86 1,458 ±3 38.7% 2026-10-02 —
Epoch AI swe_bench_verified decay ×0.37 0.74 ±0.02 1,454 ±10 13.4% 2026-05-15 —

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

A Math 1,418 [1,410–1,426] Percentile 82 · #46 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,473 ±5.85 1,463 ±6 50.4% 2026-10-02 —
Epoch AI frontiermath_t1_3 decay ×0.86 0.37 ±0.03 1,401 ±6 42.4% 2026-08-29 —

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

A Writing 1,459 [1,454–1,463] Percentile 91 · #24 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,452 ±3.37 1,482 ±3 62.0% 2026-10-02 —
LMArena text/longer_query 1,466 ±2.51 1,462 ±2 31.9% 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

A Language: 繁體中文 1,448 [1,441–1,456] Percentile 88 · #27 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,515 ±5.06 1,465 ±4 93.7% 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

A Language: Deutsch 1,437 [1,421–1,453] Percentile 85 · #23 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,464 ±10 1,457 ±9 92.2% 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

A Language: English 1,452 [1,447–1,457] Percentile 91 · #26 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,476 ±2.42 1,468 ±3 94.0% 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

A Language: Español 1,443 [1,431–1,456] Percentile 88 · #18 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,468 ±7.36 1,461 ±7 93.0% 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

A Language: Français 1,438 [1,425–1,450] Percentile 81 · #27 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,475 ±7.56 1,456 ±7 93.0% 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

A Language: 日本語 1,432 [1,416–1,449] Percentile 80 · #26 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,434 ±12 1,452 ±9 92.2% 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

A Language: 한국어 1,432 [1,419–1,446] Percentile 80 · #28 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,420 ±9.72 1,450 ±7 92.8% 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

B Language: Polski 1,428 [1,414–1,442] Percentile 74 · #35 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,445 ±9.05 1,446 ±8 92.7% 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

A Language: Русский 1,439 [1,432–1,446] Percentile 84 · #38 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,453 ±3.92 1,455 ±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

History

fedi-index by month
The same in a table
Model 2026-10
GLM 5.1 1,432 A

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: GLM 5.1

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