deepseek-v4-flash-0731

Overall

78.2

/ 100

Tasks

500/500

task suite

Avg latency

81571ms

round trip

Tokens used

75264

during evaluation

Per-category results

arithmetic 100.0 / 100

coding_light 75.0 / 100

data_manipulation 95.71 / 100

extraction 30.0 / 100

instruction_following 100.0 / 100

knowledge 85.0 / 100

language 50.0 / 100

Strongest category: language

About the suite

Fikra Eval 500 is one deterministic 500-task platform evaluation generated from a fixed seed: 110 arithmetic, 60 knowledge, 80 coding-light, 70 data manipulation, 70 language, 60 extraction and 50 instruction-following tasks. Same tasks every model, automatic scoring, published exactly when 500/500 completes.

suite :: fikra-eval-500 / v1

deepseek-v4-flash-0731 model page →

Sample task outputs (first 60 of 500)

#	Category	Result	Latency	Reply
1	arithmetic	✓	1425 ms	4195
2	arithmetic	✓	2271 ms	365
3	arithmetic	✓	2468 ms	3900
4	arithmetic	✓	1657 ms	2122
5	arithmetic	✓	2116 ms	19430
6	arithmetic	✓	4395 ms	236
7	arithmetic	✓	2007 ms	4185
8	arithmetic	✓	1536 ms	12502
9	arithmetic	✓	2825 ms	42693
10	arithmetic	✓	4112 ms	27216
11	arithmetic	✓	2900 ms	4583
12	arithmetic	✓	3061 ms	15634
13	arithmetic	✓	3550 ms	3756
14	arithmetic	✓	3392 ms	35
15	arithmetic	✓	3780 ms	326
16	arithmetic	✓	4192 ms	12029
17	arithmetic	✓	4232 ms	82
18	arithmetic	✓	3952 ms	239
19	arithmetic	✓	5391 ms	2727
20	arithmetic	✓	4706 ms	30590
21	arithmetic	✓	5551 ms	3001
22	arithmetic	✓	5071 ms	137
23	arithmetic	✓	5395 ms	1393
24	arithmetic	✓	5243 ms	23650
25	arithmetic	✓	5921 ms	4050
26	arithmetic	✓	5765 ms	5
27	arithmetic	✓	5832 ms	30848
28	arithmetic	✓	8342 ms	7341
29	arithmetic	✓	6394 ms	140
30	arithmetic	✓	7107 ms	7747
31	arithmetic	✓	6409 ms	1778
32	arithmetic	✓	7421 ms	2828
33	arithmetic	✓	6934 ms	1096
34	arithmetic	✓	6893 ms	140
35	arithmetic	✓	6881 ms	12240
36	arithmetic	✓	7466 ms	9612
37	arithmetic	✓	8149 ms	302
38	arithmetic	✓	9624 ms	5063
39	arithmetic	✓	8157 ms	3272
40	arithmetic	✓	8235 ms	334
41	arithmetic	✓	8329 ms	24960
42	arithmetic	✓	8786 ms	263
43	arithmetic	✓	8800 ms	38
44	arithmetic	✓	9767 ms	14989
45	arithmetic	✓	9487 ms	4069
46	arithmetic	✓	10841 ms	3412
47	arithmetic	✓	9689 ms	7215
48	arithmetic	✓	9586 ms	261
49	arithmetic	✓	10244 ms	108
50	arithmetic	✓	10369 ms	231
51	arithmetic	✓	10519 ms	24060
52	arithmetic	✓	11073 ms	1443
53	arithmetic	✓	10816 ms	13617
54	arithmetic	✓	11423 ms	19228
55	arithmetic	✓	11303 ms	18760
56	arithmetic	✓	12581 ms	5286
57	arithmetic	✓	11607 ms	281
58	arithmetic	✓	12670 ms	42816
59	arithmetic	✓	12910 ms	4326
60	arithmetic	✓	12845 ms	8552