Gemini 3.7 Flash in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
S 1,469 ±6
Percentile 96 · #12
People prefer / Solves tasks
1,467 / 1,471
Human votes vs benchmarks, Elo-eq
Value
+81
$1.50 per 1M tokens, median of 20 providers · vendor $1.50
On your hardware
Closed weights: cloud only.
Score decomposition
S General 1,469 [1,463–1,475] Percentile 96 · #12 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
157 ±1.15 | 1,495 ±9 | 31.9% | 2026-08-13 | — |
| LMArena |
text/overall
|
1,487 ±2.63 | 1,487 ±3 | 16.1% | 2026-10-02 | high |
| LMArena |
text/hard_prompts
|
1,496 ±3.09 | 1,481 ±3 | 16.0% | 2026-10-02 | high |
| LMArena |
text/instruction_following
|
1,483 ±3.86 | 1,492 ±4 | 7.8% | 2026-10-02 | high |
| LMArena |
text/expert
|
1,507 ±6.21 | 1,482 ±5 | 7.4% | 2026-10-02 | high |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.69 ±0.01 | 1,515 ±5 | 4.0% | 2026-08-27 | high |
| LMArena |
text/longer_query
|
1,493 ±3.57 | 1,487 ±3 | 4.0% | 2026-10-02 | high |
| LMArena |
text/multi_turn
|
1,490 ±5.57 | 1,480 ±5 | 3.7% | 2026-10-02 | high |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.95 ±0.01 | 1,481 ±6 | 1.6% | 2026-08-14 | 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 1,446 [1,443–1,450] Percentile 92 · #24 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,592 ±3.76 | 1,476 ±1 | 46.4% | 2026-10-01 | high |
| LMArena |
text/coding
|
1,498 ±4.23 | 1,469 ±4 | 43.9% | 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
C Agents 1,425 [1,422–1,428] Percentile 40 · #35 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
-0.01 ±0.00 | 1,448 ±2 | 90.4% | 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 Math 1,461 [1,451–1,470] Percentile 94 · #15 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,507 ±9.25 | 1,496 ±9 | 48.1% | 2026-10-02 | high |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.72 ±0.03 | 1,476 ±6 | 32.5% | 2026-08-14 | high |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.37 ±0.08 | 1,451 ±12 | 11.3% | 2026-08-14 | 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 1,470 [1,464–1,477] Percentile 95 · #7 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,315 ±5.32 | 1,488 ±5 | 64.7% | 2026-10-02 | high |
| LMArena |
vision/ocr
|
1,325 ±6.10 | 1,495 ±6 | 15.5% | 2026-10-02 | high |
| LMArena |
vision/diagram
|
1,329 ±10 | 1,489 ±9 | 13.1% | 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 Writing 1,490 [1,483–1,496] Percentile 97 · #9 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,491 ±4.90 | 1,521 ±5 | 61.2% | 2026-10-02 | high |
| LMArena |
text/longer_query
|
1,493 ±3.57 | 1,487 ±3 | 32.4% | 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 Language: 繁體中文 1,474 [1,463–1,485] Percentile 97 · #7 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,551 ±7.60 | 1,494 ±6 | 93.2% | 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: Deutsch 1,451 [1,426–1,476] Percentile 93 · #11 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,491 ±17 | 1,481 ±14 | 89.3% | 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 1,460 [1,453–1,467] Percentile 96 · #13 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,484 ±3.62 | 1,477 ±4 | 93.8% | 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 1,465 [1,445–1,485] Percentile 97 · #6 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,499 ±12 | 1,490 ±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: Français 1,457 [1,436–1,478] Percentile 94 · #9 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,504 ±13 | 1,482 ±12 | 90.9% | 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: 日本語 1,474 [1,452–1,497] Percentile 95 · #7 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,503 ±17 | 1,504 ±13 | 90.2% | 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: 한국어 1,471 [1,450–1,492] Percentile 98 · #3 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,483 ±16 | 1,498 ±12 | 90.7% | 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: Polski 1,467 [1,442–1,492] Percentile 95 · #7 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,507 ±17 | 1,498 ±14 | 89.4% | 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 Language: Русский 1,490 [1,480–1,500] Percentile 99 · #3 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,516 ±6.17 | 1,511 ±6 | 93.4% | 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
History
The same in a table
| Model | 2026-10 |
|---|---|
| Gemini 3.7 Flash | 1,469 S |
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.