Gemma 2 9B IT in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
D 1,188 ±9
Percentile 21 · #234
People prefer / Solves tasks
1,199 / 1,155
Human votes vs benchmarks, Elo-eq
Value
-137
$0.20 per 1M tokens, median of 1 providers
On your hardware
Closed weights: cloud only.
Score decomposition
D General 1,188 [1,179–1,196] Percentile 21 · #234 · components: 8
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,207 ±1.94 | 1,207 ±2 | 20.3% | 2026-11-11 | — |
| LMArena |
text/hard_prompts
|
1,171 ±2.99 | 1,188 ±3 | 20.0% | 2026-11-11 | — |
| Epoch AI |
eci
|
113 ±2.76 | 1,143 ±22 | 19.9% | 2024-06-24 | — |
| LMArena |
text/instruction_following
|
1,178 ±2.68 | 1,203 ±2 | 10.0% | 2026-11-11 | — |
| LMArena |
text/expert
|
1,147 ±6.03 | 1,188 ±5 | 9.3% | 2026-11-11 | — |
| LMArena |
text/multi_turn
|
1,193 ±3.72 | 1,211 ±3 | 4.9% | 2026-11-11 | — |
| LMArena |
text/longer_query
|
1,196 ±3.97 | 1,210 ±4 | 4.9% | 2026-11-11 | — |
| Epoch AI |
gpqa_diamond
decay ×0.25 |
0.27 ±0.02 | 1,168 ±11 | 1.7% | 2025-01-27 | — |
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 Coding 1,182 [1,176–1,188] Percentile 19 · #233 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,173 ±3.67 | 1,179 ±3 | 88.4% | 2026-11-11 | — |
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
D Math 1,187 [1,181–1,194] Percentile 21 · #202 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,183 ±3.92 | 1,185 ±4 | 88.3% | 2026-11-11 | — |
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
C Writing 1,222 [1,217–1,227] Percentile 26 · #196 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,205 ±3.81 | 1,230 ±4 | 62.4% | 2026-11-11 | — |
| LMArena |
text/longer_query
|
1,196 ±3.97 | 1,210 ±4 | 31.4% | 2026-11-11 | — |
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
D Language: 繁體中文 1,203 [1,196–1,210] Percentile 15 · #187 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,185 ±4.98 | 1,203 ±4 | 93.7% | 2026-11-11 | — |
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
D Language: Deutsch 1,216 [1,203–1,230] Percentile 11 · #129 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,186 ±8.35 | 1,218 ±7 | 92.9% | 2026-11-11 | — |
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
D Language: English 1,198 [1,193–1,203] Percentile 25 · #203 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,220 ±2.43 | 1,198 ±3 | 94.0% | 2026-11-11 | — |
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
D Language: Español 1,213 [1,191–1,234] Percentile 5 · #142 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,200 ±13 | 1,214 ±12 | 90.6% | 2026-11-11 | — |
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
D Language: Français 1,195 [1,173–1,217] Percentile 6 · #133 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,190 ±13 | 1,194 ±12 | 90.5% | 2026-11-11 | — |
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
D Language: 日本語 1,234 [1,221–1,246] Percentile 7 · #118 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,144 ±9.08 | 1,236 ±7 | 93.0% | 2026-11-11 | — |
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
D Language: 한국어 1,233 [1,218–1,248] Percentile 7 · #128 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,137 ±11 | 1,235 ±8 | 92.5% | 2026-11-11 | — |
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
D Language: Русский 1,226 [1,219–1,233] Percentile 16 · #193 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,200 ±4.13 | 1,227 ±4 | 93.8% | 2026-11-11 | — |
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
History
The same in a table
| Model | 2026-10 | 2026-11 |
|---|---|---|
| Gemma 2 9B IT | 1,199 | 1,188 D |
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.