DeepSeek LLM 67B Chat in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
D 1,124 ±6
Percentile 16 · #251
People prefer / Solves tasks
1,110 / 1,165
Human votes vs benchmarks, Elo-eq
Value
—
On your hardware
Closed weights: cloud only.
Score decomposition
D General 1,124 [1,118–1,130] Percentile 16 · #251 · components: 5
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/overall
|
1,105 ±5.88 | 1,105 ±6 | 27.9% | 2026-11-11 | — |
| LMArena |
text/hard_prompts
|
1,070 ±9.35 | 1,097 ±8 | 24.5% | 2026-11-11 | — |
| Epoch AI |
gpqa_diamond
decay ×0.25 |
0.25 ±0.01 | 1,154 ±6 | 18.9% | 2025-01-27 | — |
| LMArena |
text/instruction_following
|
1,079 ±8.56 | 1,109 ±8 | 12.5% | 2026-11-11 | — |
| LMArena |
text/multi_turn
|
1,082 ±12 | 1,110 ±11 | 5.2% | 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 Coding 1,124 [1,107–1,142] Percentile 13 · #249 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/coding
|
1,095 ±12 | 1,110 ±11 | 84.2% | 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 1,128 [1,109–1,146] Percentile 14 · #220 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,108 ±12 | 1,113 ±11 | 83.5% | 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 Writing 1,099 [1,078–1,119] Percentile 11 · #235 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,067 ±11 | 1,088 ±11 | 91.0% | 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: English 1,107 [1,094–1,120] Percentile 15 · #229 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,127 ±6.51 | 1,100 ±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
History
The same in a table
| Model | 2026-10 | 2026-11 |
|---|---|---|
| DeepSeek LLM 67B Chat | 1,119 | 1,124 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.