o4 Mini in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
B 1,356 ±6
Percentile 50 · #146
People prefer / Solves tasks
1,338 / 1,382
Human votes vs benchmarks, Elo-eq
Value
-37
$1.92 per 1M tokens, median of 18 providers · vendor $1.92
On your hardware
Closed weights: cloud only.
Score decomposition
B General 1,356 [1,351–1,362] Percentile 50 · #146 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
146 ±1.05 | 1,403 ±8 | 32.7% | 2025-04-16 | — |
| LMArena |
text/overall
|
1,353 ±2.02 | 1,353 ±2 | 16.0% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,351 ±2.62 | 1,350 ±2 | 15.9% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,321 ±3.08 | 1,338 ±3 | 7.8% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,344 ±6.26 | 1,348 ±5 | 7.3% | 2026-10-02 | — |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.20 ±0.01 | 1,359 ±4 | 4.1% | 2026-08-27 | low |
| LMArena |
text/longer_query
|
1,315 ±3.50 | 1,321 ±3 | 3.9% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,350 ±3.68 | 1,353 ±3 | 3.9% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.80 ±0.02 | 1,410 ±11 | 1.2% | 2025-04-16 | high |
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 1,338 [1,333–1,344] Percentile 39 · #171 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,368 ±3.50 | 1,353 ±3 | 90.2% | 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
B Math 1,377 [1,371–1,384] Percentile 55 · #115 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,390 ±5.46 | 1,383 ±5 | 50.6% | 2026-10-02 | — |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.36 ±0.03 | 1,399 ±6 | 27.9% | 2026-06-11 | high |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.05 ±0.03 | 1,400 ±5 | 14.4% | 2026-06-11 | high |
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 1,366 [1,361–1,370] Percentile 35 · #74 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,194 ±3.60 | 1,379 ±3 | 63.8% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,197 ±4.54 | 1,372 ±4 | 15.4% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,197 ±6.47 | 1,372 ±6 | 14.6% | 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 1,313 [1,307–1,318] Percentile 40 · #161 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,294 ±4.18 | 1,320 ±4 | 61.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,315 ±3.50 | 1,321 ±3 | 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 1,329 [1,321–1,338] Percentile 33 · #149 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,355 ±5.61 | 1,338 ±4 | 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
C Language: Deutsch 1,337 [1,323–1,351] Percentile 29 · #102 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,336 ±8.90 | 1,347 ±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
C Language: English 1,344 [1,340–1,350] Percentile 44 · #152 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,367 ±2.58 | 1,354 ±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 1,338 [1,321–1,354] Percentile 16 · #125 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,347 ±9.93 | 1,349 ±9 | 92.1% | 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
D Language: Français 1,338 [1,318–1,359] Percentile 16 · #113 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,362 ±13 | 1,352 ±12 | 90.9% | 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: 日本語 1,346 [1,333–1,360] Percentile 29 · #91 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,308 ±9.82 | 1,358 ±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
C Language: 한국어 1,357 [1,343–1,371] Percentile 30 · #95 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,313 ±10 | 1,369 ±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
D Language: Polski 1,361 [1,352–1,370] Percentile 22 · #103 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,358 ±6.04 | 1,372 ±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: Русский 1,338 [1,330–1,347] Percentile 38 · #145 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,334 ±5.02 | 1,348 ±4 | 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 |
|---|---|
| o4 Mini | 1,356 B |
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.