gemini-3-pro in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,448 ±6
Percentile 89 · #32
People prefer / Solves tasks
1,454 / 1,440
Human votes vs benchmarks, Elo-eq
Value
+33
$4.50 per 1M tokens, median of 6 providers
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,448 [1,442–1,454] Percentile 89 · #32 · components: 8
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
153 ±1.06 | 1,461 ±8 | 36.8% | 2025-11-18 | — |
| LMArena |
text/overall
|
1,479 ±1.92 | 1,479 ±2 | 15.9% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,481 ±2.47 | 1,468 ±2 | 15.8% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,458 ±3.18 | 1,468 ±3 | 7.8% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,475 ±5.70 | 1,456 ±5 | 7.4% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,471 ±3.25 | 1,467 ±3 | 3.9% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,484 ±3.95 | 1,475 ±4 | 3.8% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.93 ±0.02 | 1,470 ±8 | 1.7% | 2025-11-19 | — |
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 1,424 [1,420–1,427] Percentile 80 · #56 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,439 ±4.29 | 1,432 ±1 | 39.8% | 2026-10-01 | — |
| LMArena |
text/coding
|
1,482 ±3.52 | 1,454 ±3 | 38.4% | 2026-10-02 | — |
| Epoch AI |
swe_bench_verified
decay ×0.37 |
0.73 ±0.02 | 1,447 ±10 | 13.4% | 2026-02-13 | — |
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 Agents 1,396 [1,386–1,407] Percentile 12 · #51 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| τ²-bench (Sierra) |
tau2/banking_knowledge
decay ×0.76 |
0.18 ±0.04 | 1,444 ±6 | 80.4% | 2026-03-02 | 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
A Math 1,433 [1,424–1,442] Percentile 86 · #36 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,475 ±5.67 | 1,466 ±5 | 87.6% | 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
A Vision 1,460 [1,455–1,465] Percentile 89 · #13 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,305 ±3.77 | 1,479 ±3 | 63.4% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,310 ±4.21 | 1,481 ±4 | 15.5% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,301 ±5.96 | 1,464 ±5 | 14.8% | 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
A Writing 1,478 [1,473–1,484] Percentile 96 · #12 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,482 ±4.12 | 1,512 ±4 | 61.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,471 ±3.25 | 1,467 ±3 | 32.0% | 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: 繁體中文 1,454 [1,446–1,462] Percentile 92 · #19 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,522 ±5.30 | 1,471 ±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 1,479 [1,461–1,496] Percentile 99 · #2 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,517 ±11 | 1,503 ±10 | 91.9% | 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: English 1,454 [1,448–1,459] Percentile 92 · #24 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,478 ±2.76 | 1,470 ±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
A Language: Español 1,442 [1,427–1,457] Percentile 86 · #22 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,468 ±8.95 | 1,461 ±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
A Language: Français 1,450 [1,434–1,467] Percentile 90 · #15 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,492 ±10 | 1,472 ±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
A Language: 日本語 1,484 [1,463–1,504] Percentile 99 · #2 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,513 ±15 | 1,511 ±11 | 91.1% | 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
A Language: 한국어 1,451 [1,435–1,467] Percentile 90 · #15 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,449 ±12 | 1,472 ±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: Polski 1,474 [1,461–1,487] Percentile 98 · #3 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,503 ±8.41 | 1,494 ±7 | 92.9% | 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: Русский 1,472 [1,465–1,479] Percentile 96 · #11 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,492 ±4.37 | 1,490 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| gemini-3-pro | 1,448 A |
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.