deepseek-coder-v2 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
D 1,207 ±4
Percentile 24 · #224 · provisional
People prefer / Solves tasks
1,207 / —
Human votes vs benchmarks, Elo-eq
Value
—
On your hardware
Closed weights: cloud only.
Score decomposition
D General 1,207 [1,203–1,211] Percentile 24 · #224 · components: 6 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,191 ±3.23 | 1,191 ±3 | 26.8% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,207 ±4.99 | 1,220 ±4 | 25.7% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,180 ±4.63 | 1,205 ±4 | 12.9% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,181 ±11 | 1,216 ±9 | 10.4% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,176 ±6.34 | 1,196 ±6 | 6.1% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,219 ±7.09 | 1,231 ±7 | 5.9% | 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
C Coding 1,244 [1,234–1,253] Percentile 26 · #205 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,251 ±6.06 | 1,248 ±5 | 89.5% | 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
C Math 1,236 [1,225–1,247] Percentile 28 · #183 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,241 ±6.92 | 1,241 ±7 | 87.0% | 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
D Writing 1,164 [1,154–1,174] Percentile 21 · #212 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,120 ±6.81 | 1,143 ±7 | 73.1% | 2026-10-02 | — |
| LMArena |
text/longer_query
outlier |
1,219 ±7.09 | 1,231 ±7 | 18.6% | 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
D Language: 繁體中文 1,215 [1,204–1,226] Percentile 18 · #184 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,201 ±7.70 | 1,216 ±6 | 93.2% | 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
D Language: Deutsch 1,199 [1,179–1,220] Percentile 15 · #123 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,164 ±13 | 1,199 ±12 | 90.9% | 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
D Language: English 1,173 [1,165–1,181] Percentile 22 · #212 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,194 ±4.10 | 1,171 ±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
D Language: 日本語 1,220 [1,199–1,242] Percentile 10 · #115 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,126 ±16 | 1,222 ±12 | 90.6% | 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
D Language: 한국어 1,209 [1,188–1,231] Percentile 10 · #122 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,104 ±16 | 1,210 ±12 | 90.5% | 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
D Language: Русский 1,215 [1,203–1,227] Percentile 16 · #197 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,188 ±7.19 | 1,216 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| deepseek-coder-v2 | 1,207 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.