GPT-4.1 Mini in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
C 1,327 ±6
Percentile 40 · #175
People prefer / Solves tasks
1,337 / 1,311
Human votes vs benchmarks, Elo-eq
Value
-42
$0.70 per 1M tokens, median of 23 providers · vendor $0.70
On your hardware
Closed weights: cloud only.
Score decomposition
C General 1,327 [1,321–1,334] Percentile 40 · #175 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
135 ±1.37 | 1,320 ±11 | 29.6% | 2025-04-14 | — |
| LMArena |
text/overall
|
1,340 ±2.20 | 1,340 ±2 | 16.8% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,349 ±2.88 | 1,348 ±3 | 16.7% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,332 ±3.31 | 1,349 ±3 | 8.2% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,338 ±6.62 | 1,344 ±5 | 7.6% | 2026-10-02 | — |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.13 ±0.01 | 1,337 ±3 | 4.4% | 2026-08-31 | — |
| LMArena |
text/multi_turn
|
1,354 ±3.88 | 1,357 ±4 | 4.1% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,344 ±3.82 | 1,348 ±4 | 4.1% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.66 ±0.03 | 1,346 ±13 | 1.2% | 2025-04-14 | — |
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 1,337 [1,331–1,343] Percentile 38 · #172 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,367 ±3.83 | 1,352 ±3 | 90.1% | 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
C Math 1,328 [1,322–1,334] Percentile 38 · #158 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,343 ±5.67 | 1,338 ±5 | 48.2% | 2026-10-02 | — |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.07 ±0.01 | 1,336 ±3 | 45.0% | 2026-08-27 | — |
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 Vision 1,355 [1,350–1,360] Percentile 30 · #79 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,181 ±3.81 | 1,368 ±3 | 63.9% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,188 ±4.81 | 1,363 ±5 | 15.4% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,182 ±6.95 | 1,359 ±6 | 14.4% | 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 Writing 1,325 [1,319–1,331] Percentile 44 · #151 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,300 ±4.45 | 1,326 ±4 | 61.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,344 ±3.82 | 1,348 ±4 | 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
C Language: 繁體中文 1,310 [1,301–1,319] Percentile 30 · #157 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,329 ±6.15 | 1,318 ±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
C Language: Deutsch 1,347 [1,332–1,361] Percentile 32 · #98 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,349 ±9.36 | 1,358 ±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 1,333 [1,327–1,338] Percentile 40 · #164 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,356 ±2.76 | 1,341 ±3 | 93.9% | 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: Español 1,314 [1,296–1,332] Percentile 13 · #130 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,320 ±11 | 1,324 ±10 | 91.8% | 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: Français 1,334 [1,313–1,355] Percentile 16 · #114 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,357 ±13 | 1,347 ±12 | 90.7% | 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
C Language: 日本語 1,335 [1,321–1,348] Percentile 25 · #96 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,291 ±10 | 1,345 ±8 | 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
C Language: 한국어 1,346 [1,331–1,361] Percentile 25 · #102 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,298 ±11 | 1,357 ±8 | 92.4% | 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: Polski 1,335 [1,326–1,345] Percentile 14 · #113 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,326 ±6.22 | 1,345 ±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
C Language: Русский 1,330 [1,321–1,338] Percentile 34 · #155 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,324 ±5.38 | 1,338 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| GPT-4.1 Mini | 1,327 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.