Skip to content
fedi.software

GLM 4.7 Flash in fedi-index

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

fedi-index · General

C 1,332 ±4

Percentile 43 · #167

People prefer / Solves tasks

1,342 / 1,286

Human votes vs benchmarks, Elo-eq

Value

+1

$0.1489 per 1M tokens, median of 12 providers

On your hardware

  • GPU 12 GB #15 · Q8_0 · ≈11 tok/s
  • GPU 24 GB #13 · UD-Q5_K_XL · ≈103 tok/s
  • Mac 64 GB #12 · Q8_0 · ≈32 tok/s
  • 128 GB unified memory #19 · Q8_0 · ≈23 tok/s
  • CPU, 32 GB RAM #6 · Q4_K_M · ≈6 tok/s

Score decomposition

C General 1,332 [1,328–1,336] Percentile 43 · #167 · components: 7
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,351 ±2.90 1,351 ±3 23.2% 2026-10-02 —
LMArena text/hard_prompts 1,356 ±3.72 1,355 ±3 23.0% 2026-10-02 —
Epoch AI gpqa_diamond decay ×0.26 0.61 ±0.03 1,321 ±14 12.2% 2026-08-30 —
LMArena text/instruction_following 1,327 ±5.17 1,344 ±5 10.9% 2026-10-02 —
LMArena text/expert 1,357 ±9.74 1,359 ±8 9.4% 2026-10-02 —
LMArena text/longer_query 1,345 ±5.15 1,349 ±5 5.5% 2026-10-02 —
LMArena text/multi_turn 1,342 ±6.52 1,346 ±6 5.2% 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

C Coding 1,349 [1,340–1,358] Percentile 44 · #155 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,383 ±5.57 1,366 ±5 89.6% 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

C Math 1,327 [1,310–1,344] Percentile 38 · #159 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,354 ±11 1,349 ±10 84.6% 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

C Writing 1,324 [1,315–1,333] Percentile 43 · #152 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,298 ±6.97 1,324 ±7 59.9% 2026-10-02 —
LMArena text/longer_query 1,345 ±5.15 1,349 ±5 33.3% 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

C Language: 繁體中文 1,363 [1,348–1,378] Percentile 42 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,403 ±10 1,376 ±8 92.5% 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

C Language: English 1,350 [1,341–1,358] Percentile 46 · #149 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,373 ±4.34 1,360 ±5 93.6% 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 Language: Español 1,338 [1,314–1,362] Percentile 17 · #124 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,351 ±14 1,353 ±13 89.9% 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: Polski 1,302 [1,278–1,327] Percentile 3 · #127 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,290 ±17 1,314 ±14 89.4% 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

C Language: Русский 1,337 [1,325–1,348] Percentile 36 · #149 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,333 ±7.10 1,347 ±6 93.1% 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

History

fedi-index by month
The same in a table
Model 2026-10
GLM 4.7 Flash 1,332 C

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 4.7 Flash

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