Gemini 3.1 Pro Preview in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,457 ±6
Percentile 93 · #21
People prefer / Solves tasks
1,458 / 1,456
Human votes vs benchmarks, Elo-eq
Value
+42
$4.50 per 1M tokens, median of 28 providers · vendor $4.50
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,457 [1,451–1,463] Percentile 93 · #21 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
155 ±1.23 | 1,475 ±10 | 30.5% | 2026-02-19 | — |
| LMArena |
text/overall
|
1,480 ±1.53 | 1,480 ±2 | 16.3% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,484 ±1.86 | 1,471 ±2 | 16.3% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,465 ±2.22 | 1,475 ±2 | 8.1% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,483 ±3.24 | 1,462 ±3 | 8.0% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,482 ±2.16 | 1,477 ±2 | 4.0% | 2026-10-02 | — |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.74 ±0.01 | 1,528 ±4 | 4.0% | 2026-08-10 | high |
| LMArena |
text/multi_turn
|
1,487 ±2.73 | 1,478 ±2 | 4.0% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.94 ±0.02 | 1,479 ±8 | 1.5% | 2026-08-06 | 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
B Coding 1,422 [1,420–1,424] Percentile 79 · #60 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,446 ±2.57 | 1,434 ±1 | 45.6% | 2026-10-01 | — |
| LMArena |
text/coding
|
1,484 ±2.36 | 1,456 ±2 | 44.8% | 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 Agents 1,414 [1,409–1,419] Percentile 33 · #39 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
-0.08 ±0.00 | 1,419 ±2 | 65.6% | 2026-10-02 | — |
| τ²-bench (Sierra) |
tau2/banking_knowledge
decay ×0.76 |
0.26 ±0.04 | 1,457 ±8 | 27.3% | 2026-05-05 | 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 1,444 [1,438–1,450] Percentile 90 · #27 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,485 ±4.16 | 1,474 ±4 | 53.9% | 2026-10-02 | — |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.60 ±0.03 | 1,450 ±6 | 28.2% | 2026-06-11 | — |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.27 ±0.07 | 1,435 ±11 | 10.7% | 2026-06-11 | — |
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 1,456 [1,452–1,459] Percentile 85 · #18 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,296 ±2.66 | 1,471 ±2 | 62.8% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,303 ±2.58 | 1,474 ±2 | 15.7% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,311 ±3.58 | 1,473 ±3 | 15.4% | 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
A Writing 1,482 [1,479–1,486] Percentile 97 · #10 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,482 ±2.77 | 1,512 ±3 | 62.3% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,482 ±2.16 | 1,477 ±2 | 31.7% | 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: 繁體中文 1,459 [1,454–1,465] Percentile 94 · #14 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,528 ±3.80 | 1,476 ±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: Deutsch 1,462 [1,450–1,474] Percentile 97 · #5 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,492 ±7.49 | 1,482 ±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
A Language: English 1,455 [1,451–1,459] Percentile 93 · #21 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,479 ±2.03 | 1,471 ±2 | 94.0% | 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: Español 1,454 [1,444–1,463] Percentile 95 · #9 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,479 ±5.58 | 1,471 ±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
A Language: Français 1,449 [1,440–1,459] Percentile 89 · #16 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,487 ±5.90 | 1,467 ±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
A Language: 日本語 1,477 [1,464–1,490] Percentile 98 · #4 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,495 ±9.20 | 1,498 ±7 | 93.0% | 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: 한국어 1,461 [1,450–1,471] Percentile 95 · #8 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,458 ±7.50 | 1,479 ±6 | 93.3% | 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
A Language: Polski 1,469 [1,459–1,479] Percentile 97 · #5 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,495 ±6.46 | 1,488 ±6 | 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
A Language: Русский 1,476 [1,470–1,481] Percentile 97 · #9 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,496 ±3.09 | 1,493 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| Gemini 3.1 Pro Preview | 1,457 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.