GLM-4 (0520) in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
C 1,222 ±4
Percentile 26 · #220 · provisional
People prefer / Solves tasks
1,223 / —
Human votes vs benchmarks, Elo-eq
Value
—
On your hardware
Closed weights: cloud only.
Score decomposition
C General 1,222 [1,218–1,226] Percentile 26 · #220 · components: 6 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,226 ±3.53 | 1,226 ±4 | 27.3% | 2026-11-11 | 0520 |
| LMArena |
text/hard_prompts
|
1,212 ±5.64 | 1,225 ±5 | 25.8% | 2026-11-11 | 0520 |
| LMArena |
text/instruction_following
|
1,203 ±5.20 | 1,226 ±5 | 13.0% | 2026-11-11 | 0520 |
| LMArena |
text/expert
|
1,179 ±13 | 1,214 ±10 | 9.8% | 2026-11-11 | 0520 |
| LMArena |
text/multi_turn
|
1,218 ±7.46 | 1,234 ±7 | 6.0% | 2026-11-11 | 0520 |
| LMArena |
text/longer_query
|
1,207 ±8.47 | 1,220 ±8 | 5.6% | 2026-11-11 | 0520 |
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 1,224 [1,214–1,235] Percentile 24 · #218 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,228 ±7.10 | 1,228 ±6 | 87.2% | 2026-11-11 | 0520 |
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 1,216 [1,204–1,229] Percentile 26 · #191 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,218 ±7.92 | 1,218 ±8 | 86.4% | 2026-11-11 | 0520 |
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 1,220 [1,209–1,230] Percentile 25 · #199 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,198 ±7.69 | 1,222 ±8 | 61.8% | 2026-11-11 | 0520 |
| LMArena |
text/longer_query
|
1,207 ±8.47 | 1,220 ±8 | 30.8% | 2026-11-11 | 0520 |
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 1,235 [1,223–1,248] Percentile 21 · #174 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,229 ±8.62 | 1,238 ±7 | 93.0% | 2026-11-11 | 0520 |
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 1,214 [1,191–1,238] Percentile 10 · #130 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,184 ±15 | 1,216 ±13 | 90.0% | 2026-11-11 | 0520 |
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 1,220 [1,211–1,229] Percentile 28 · #195 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,242 ±4.66 | 1,221 ±5 | 93.5% | 2026-11-11 | 0520 |
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 1,168 [1,143–1,192] Percentile 4 · #132 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,042 ±19 | 1,164 ±14 | 89.5% | 2026-11-11 | 0520 |
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 1,236 [1,223–1,250] Percentile 18 · #190 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,213 ±8.24 | 1,239 ±7 | 92.8% | 2026-11-11 | 0520 |
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
The same in a table
| Model | 2026-11 |
|---|---|
| GLM-4 (0520) | 1,222 C |
Badge
Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.
Data CC BY 4.0: keep the link to fedi.software.