Gemini 3.6 Flash in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,456 ±5
Percentile 91 · #26
People prefer / Solves tasks
1,458 / 1,453
Human votes vs benchmarks, Elo-eq
Value
+68
$1.50 per 1M tokens, median of 22 providers · vendor $1.50
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,456 [1,450–1,461] Percentile 91 · #26 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
154 ±0.93 | 1,471 ±7 | 33.8% | 2026-07-21 | — |
| LMArena |
text/overall
|
1,479 ±2.24 | 1,480 ±2 | 15.6% | 2026-10-02 | high |
| LMArena |
text/hard_prompts
|
1,486 ±2.63 | 1,472 ±2 | 15.6% | 2026-10-02 | high |
| LMArena |
text/instruction_following
|
1,466 ±3.20 | 1,476 ±3 | 7.7% | 2026-10-02 | high |
| LMArena |
text/expert
|
1,489 ±4.88 | 1,467 ±4 | 7.4% | 2026-10-02 | high |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.66 ±0.01 | 1,505 ±5 | 3.9% | 2026-08-27 | high |
| LMArena |
text/longer_query
|
1,475 ±3.02 | 1,471 ±3 | 3.8% | 2026-10-02 | high |
| LMArena |
text/multi_turn
|
1,481 ±4.32 | 1,473 ±4 | 3.7% | 2026-10-02 | high |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.94 ±0.01 | 1,478 ±6 | 1.6% | 2026-08-02 | 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 Coding 1,436 [1,434–1,439] Percentile 88 · #35 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,536 ±3.81 | 1,460 ±1 | 46.0% | 2026-10-01 | high |
| LMArena |
text/coding
|
1,492 ±3.46 | 1,463 ±3 | 44.4% | 2026-10-02 | 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
D Agents 1,398 [1,393–1,402] Percentile 14 · #50 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
-0.08 ±0.01 | 1,419 ±3 | 90.3% | 2026-10-02 | 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 1,450 [1,442–1,458] Percentile 91 · #24 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,504 ±7.25 | 1,493 ±7 | 50.5% | 2026-10-02 | high |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.59 ±0.03 | 1,449 ±6 | 30.0% | 2026-08-02 | high |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.22 ±0.07 | 1,428 ±10 | 11.9% | 2026-08-02 | 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 Vision 1,456 [1,451–1,462] Percentile 86 · #17 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,297 ±4.29 | 1,472 ±4 | 63.7% | 2026-10-02 | high |
| LMArena |
vision/ocr
|
1,309 ±4.66 | 1,480 ±4 | 15.6% | 2026-10-02 | high |
| LMArena |
vision/diagram
|
1,312 ±7.23 | 1,474 ±6 | 14.4% | 2026-10-02 | 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 Writing 1,470 [1,464–1,475] Percentile 94 · #16 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,466 ±4.07 | 1,496 ±4 | 61.7% | 2026-10-02 | high |
| LMArena |
text/longer_query
|
1,475 ±3.02 | 1,471 ±3 | 32.1% | 2026-10-02 | 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 Language: 繁體中文 1,460 [1,451–1,469] Percentile 95 · #13 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,531 ±6.06 | 1,478 ±5 | 93.6% | 2026-10-02 | 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 Language: Deutsch 1,444 [1,425–1,464] Percentile 90 · #16 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,476 ±13 | 1,468 ±11 | 91.3% | 2026-10-02 | 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 Language: English 1,458 [1,452–1,464] Percentile 94 · #16 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,482 ±2.99 | 1,474 ±3 | 93.9% | 2026-10-02 | 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 Language: Español 1,447 [1,430–1,465] Percentile 90 · #16 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,476 ±10 | 1,469 ±10 | 91.9% | 2026-10-02 | 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 Language: Français 1,460 [1,444–1,476] Percentile 96 · #7 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,503 ±9.72 | 1,482 ±9 | 92.2% | 2026-10-02 | 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 Language: 日本語 1,460 [1,442–1,478] Percentile 91 · #13 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,476 ±13 | 1,484 ±10 | 91.7% | 2026-10-02 | 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 Language: 한국어 1,443 [1,427–1,460] Percentile 87 · #19 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,438 ±12 | 1,464 ±9 | 92.2% | 2026-10-02 | 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 Language: Polski 1,451 [1,431–1,471] Percentile 90 · #14 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,480 ±13 | 1,475 ±11 | 91.1% | 2026-10-02 | 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 Language: Русский 1,468 [1,460–1,476] Percentile 94 · #14 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,488 ±4.96 | 1,486 ±4 | 93.6% | 2026-10-02 | 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
History
The same in a table
| Model | 2026-10 |
|---|---|
| Gemini 3.6 Flash | 1,456 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.