Models tracked
—
Across the selected period
Community telemetry
See how models perform in the real world.
Usage, speed, value, and coding benchmarks—grounded in how CodingFleet users actually work, not marketing claims.
Tokens served
—
Prompt and completion tokens
Fastest response
—
Characters per second
Community favorite
—
Models with 10+ votes
Token usage over time
Model usage across CodingFleet.
Loading usage data…
Model leaderboard
Sort any metric, then narrow the list without reloading.
| Model | Speed | Usage | Credits | User score | Coding | LiveBench | WebDev Arena |
|---|---|---|---|---|---|---|---|
1
GPT-4 Turbo
|
43.7 chars/s | 62 31 | 10 | — | — | — | — |
2
Grok-3 (Beta)
|
99.0 chars/s | 142 3 | 3 | — | 73.6 | 63.2 | — |
3
Claude 3.5 Sonnet
|
167.6 chars/s | 319 77 | 2 | — | — | — | — |
4
DeepSeek V3
|
204.1 chars/s | 1.1K 9 | 1 | — | — | — | — |
|
5
DeepSeek R1
|
273.3 chars/s | 1.4K 1 | 1 | — | — | — | — |
6
Llama 3.3 70B
|
136.6 chars/s | 1.5K 19 | 1 | — | 24.1 | 45.7 | — |
|
7
GPT-3.5 Turbo
|
85.0 chars/s | 10.2K 56 | 1 | — | — | — | — |
8
Qwen2.5-Coder 32B
|
357.2 chars/s | 31.9K 8 | 1 | — | 56.9 | 46.2 | 904.1 |
|
9
GPT-4o Mini
|
16.6 chars/s | 184.9K 65 | 1 | — | 43.2 | 41.3 | — |
|
10
Grok 4.5
|
76.3 chars/s | 248.7K 24 | 4 | — | — | — | — |
|
11
Llama 3.1 405B
|
105.5 chars/s | 261.8K 17 | 2 | — | 42.7 | 52.4 | 813.7 |
12
Mistral Large 3
|
253.3 chars/s | 823.1K 25 | 1 | — | 62.9 | 50.3 | — |
|
13
Grok-4.3
|
349.3 chars/s | 1.4M 25 | 3 | — | — | — | — |
14
Kimi K2 (0905)
|
184.7 chars/s | 1.6M 25 | 1 | — | 71.8 | 62.7 | — |
|
15
Grok-3 Mini (Beta)
|
294.3 chars/s | 1.7M 8 | 1 | — | 54.5 | 70.3 | — |
16
Gemini 3.6 Flash
|
389.6 chars/s | 2.0M 26 | 4 | — | — | — | — |
|
17
Qwen 3.7 Plus
|
319.5 chars/s | 2.6M 26 | 1 | — | — | — | — |
|
18
Grok-4.20
|
700.7 chars/s | 2.8M 25 | 4 | — | — | — | — |
|
19
Kimi K2 Thinking
|
138.5 chars/s | 3.0M 25 | 1 | — | — | — | — |
|
20
GPT-4o
|
223.5 chars/s | 3.0M 52 | 2 | — | — | — | — |
|
21
Qwen3 Coder
|
293.2 chars/s | 3.3M 24 | 2 | — | 73.2 | 60.5 | — |
22
Gemma 4 26B
|
62.8 chars/s | 4.1M 24 | 0 | — | — | — | — |
|
23
GPT-5.2 Codex
|
225.7 chars/s | 5.2M 22 | 8 | — | — | — | — |
|
24
Codestral (2508)
|
433.2 chars/s | 5.2M 22 | 1 | — | — | — | — |
|
25
Mistral Medium 3.5
|
374.6 chars/s | 5.6M 22 | 2 | — | — | — | — |
|
26
Llama 4 Maverick
|
288.9 chars/s | 6.1M 4 | 1 | — | 54.2 | 55.2 | 998.5 |
27
GLM 4.7
|
91.8 chars/s | 7.2M 22 | 1 | — | 73.1 | 58.1 | — |
|
28
Mistral Medium 3
|
264.0 chars/s | 7.9M 20 | 1 | — | 61.5 | 56.6 | 1160.1 |
|
29
Claude 4 Opus
|
124.5 chars/s | 8.2M | 50–60 | — | 73.3 | 72.9 | 1405.5 |
|
30
Kimi K3
|
100.7 chars/s | 10.4M 21 | 7 | — | — | — | — |
|
31
GPT-5 Pro
|
35.0 chars/s | 12.9M 21 | 70 | — | 72.1 | 78.7 | — |
|
32
GLM 4.6
|
159.0 chars/s | 13.0M 21 | 1 | — | — | — | — |
|
33
Grok-4.20 Beta
|
627.6 chars/s | 13.6M 21 | 4 | — | — | — | — |
|
34
Grok-4
|
126.8 chars/s | 15.2M 19 | 5 | — | 71.3 | 72.1 | — |
|
35
Qwen 3.6 Max Preview
|
50.5 chars/s | 16.8M 19 | 4 | — | — | — | — |
|
36
OpenAI o4-mini
|
157.7 chars/s | 17.5M 25 | 2 | — | — | — | 1095.1 |
37
MiMo V2 Pro
|
302.5 chars/s | 18.6M 20 | 1 | — | — | — | — |
38
Hy3
|
177.1 chars/s | 20.1M 19 | 1 | — | — | — | — |
|
39
Gemini 2.0 Flash
|
530.6 chars/s | 22.9M 28 | 1 | — | 64.7 | 60.1 | 1035.2 |
|
40
Qwen3 235B A22B
|
159.2 chars/s | 23.8M 19 | 1 | — | 66.4 | 64.9 | — |
|
41
Claude 3.7 Sonnet
|
176.1 chars/s | 27.9M 35 | 6–7 | — | 74.3 | 67.4 | 1356.7 |
|
42
DeepSeek V3.2 Thinking
|
49.7 chars/s | 30.5M 4 | 1 | — | 64.6 | 66.6 | — |
|
43
Grok Build 0.1
|
326.5 chars/s | 39.1M 20 | 1 | — | — | — | — |
44
MiniMax M2.5
|
143.4 chars/s | 39.5M 20 | 1 | — | 70.7 | 60.1 | — |
|
45
GLM 5V Turbo
|
223.6 chars/s | 54.8M 15 | 1 | — | — | — | — |
|
46
MiMo V2.5 Pro
|
191.1 chars/s | 56.3M 15 | 1 | — | — | — | — |
|
47
Qwen 3.6 Flash
|
234.1 chars/s | 61.0M 14 | 1 | — | — | — | — |
|
48
Kimi K2.7 Code
|
176.8 chars/s | 62.1M 14 | 1 | — | — | — | — |
|
49
Qwen 3.5 Plus
|
178.5 chars/s | 63.5M 14 | 1 | — | — | — | — |
|
50
GPT-5.6 Luna
|
428.1 chars/s | 66.1M 14 | 1–5 | — | — | — | — |
|
51
Qwen 3.5 Flash
|
339.4 chars/s | 67.7M 14 | 1 | — | — | — | — |
|
52
Kimi K2.6
|
62.5 chars/s | 71.0M 13 | 1 | — | — | — | — |
|
53
OpenAI o3
|
132.9 chars/s | 76.6M 9 | 4–5 | — | 77.9 | 74.6 | 1188.3 |
|
54
Grok-4 Fast
|
373.6 chars/s | 76.7M 13 | 1 | 0.4 | — | — | — |
|
55
Grok-4.1 Fast
|
165.3 chars/s | 80.1M 13 | 1 | — | — | — | — |
|
56
GLM 5 Turbo
|
164.2 chars/s | 86.2M 13 | 1 | — | — | — | — |
|
57
GPT-5.1 Codex Max
|
196.7 chars/s | 98.6M 13 | 4–5 | — | 81.4 | 75.2 | — |
|
58
GLM 5
|
186.2 chars/s | 120.2M 11 | 1 | — | 73.6 | 68.9 | 1456.0 |
|
59
GPT-5.6 Sol
|
351.7 chars/s | 126.9M 11 | 14–24 | 1.0 | — | — | — |
|
60
GLM 5.2
|
253.3 chars/s | 130.9M 11 | 2 | — | — | — | — |
|
61
Gemini 3.5 Flash
|
309.6 chars/s | 133.6M 11 | 4 | 1.0 | — | — | — |
|
62
MiniMax M3
|
275.9 chars/s | 135.5M 11 | 1 | — | — | — | — |
|
63
Qwen 3.7 Max
|
134.9 chars/s | 138.0M 11 | 3 | — | — | — | — |
|
64
Claude 4.1 Opus
|
113.8 chars/s | 138.5M 11 | 50–60 | — | 74.0 | 73.5 | 1476.5 |
|
65
Kimi K2.5
|
94.8 chars/s | 140.6M 11 | 1 | — | 77.9 | 69.1 | — |
|
66
Claude Fable 5 Thinking
|
154.0 chars/s | 147.5M 10 | 34 | — | — | — | — |
|
67
GPT-5.6 Terra
|
344.5 chars/s | 150.2M 10 | 7–12 | — | — | — | — |
|
68
MiniMax M2.7
|
193.5 chars/s | 170.3M 10 | 1 | — | — | — | — |
|
69
Qwen 3.6 Plus
|
73.0 chars/s | 171.5M 9 | 1 | — | — | — | — |
|
70
Claude 3.5 Haiku
|
204.1 chars/s | 185.3M 14 | 1 | — | 53.2 | 45.0 | 1133.8 |
|
71
GPT-4.1 mini
|
211.7 chars/s | 205.3M | 1 | 0.0 | 72.1 | 59.1 | 1194.4 |
|
72
GLM 5.1
|
133.4 chars/s | 241.0M 7 | 1 | — | — | — | — |
|
73
GPT-4.1
|
200.1 chars/s | 285.6M 4 | 6 | — | 73.2 | 63.0 | 1256.5 |
|
74
Claude Sonnet 5
|
196.9 chars/s | 288.4M 8 | 6–7 | — | — | — | — |
|
75
Gemini 2.5 Pro
|
166.3 chars/s | 295.7M 6 | 4 | 0.6 | 72.9 | 79.0 | 1409.1 |
|
76
Gemini 2.5 Flash
|
360.7 chars/s | 332.7M 8 | 1 | 0.3 | 60.3 | 69.9 | 1304.9 |
|
77
Gemini 3 Pro
|
160.0 chars/s | 411.0M 7 | 4 | 1.0 | 74.6 | 79.7 | 1481.0 |
|
78
GPT-5.3 Codex
|
109.9 chars/s | 497.3M 5 | 5–9 | — | — | — | — |
|
79
DeepSeek V3.2
|
98.3 chars/s | 498.0M 3 | 1 | 0.3 | 68.5 | 62.6 | — |
|
80
GPT-5.1 Codex Mini
|
262.6 chars/s | 550.4M 5 | 1 | 0.7 | 71.8 | 69.8 | 1248.0 |
|
81
Gemini 3.1 Pro
|
177.5 chars/s | 592.6M 3 | 4 | 1.0 | 76.5 | 79.9 | 1500.0 |
|
82
GPT-5.5
|
207.9 chars/s | 602.2M 3 | 13–22 | — | 82.5 | 80.7 | — |
|
83
GPT-5.2 Codex
|
111.8 chars/s | 646.4M 2 | 5–9 | 0.8 | 83.6 | 74.3 | — |
|
84
GPT-5.1
|
167.6 chars/s | 707.0M 1 | 3–5 | 0.9 | 72.5 | 72.5 | 1395.0 |
|
85
Claude Opus 4.8
|
149.5 chars/s | 1.0B | 15–17 | — | — | — | — |
|
86
Claude Opus 4.7
|
139.8 chars/s | 1.0B | 15–17 | — | 82.1 | 76.9 | 1503.0 |
|
87
Claude 4 Sonnet
|
162.3 chars/s | 1.2B 2 | 6–7 | 0.7 | 77.5 | 72.1 | 1381.8 |
|
88
GPT-5.4 Mini
|
343.4 chars/s | 1.5B 1 | 1–2 | 0.1 | — | — | — |
|
89
GPT-5
|
93.0 chars/s | 1.6B 1 | 3–5 | 1.0 | 75.3 | 78.6 | 1480.5 |
|
90
Claude Opus 4.5
|
177.5 chars/s | 2.0B | 15–17 | — | 79.7 | 79.8 | 1518.0 |
|
91
Gemini 3 Flash
|
299.2 chars/s | 2.0B | 1 | 0.2 | 73.9 | 73.6 | 1465.0 |
|
92
GPT-5 Mini
|
139.2 chars/s | 2.2B | 1–2 | 0.7 | 66.4 | 72.2 | — |
|
93
GPT-5.4
|
144.6 chars/s | 2.2B | 6–11 | 0.9 | 77.5 | 80.3 | 1481.0 |
|
94
Claude Opus 4.6
|
129.1 chars/s | 2.6B | 15–17 | 1.0 | 78.2 | 76.3 | 1547.0 |
|
95
GPT-5.2
|
117.4 chars/s | 2.9B | 4–8 | 0.8 | 76.1 | 73.6 | 1485.0 |
|
96
Claude Sonnet 4.6
|
145.0 chars/s | 3.2B | 9–10 | 1.0 | 80.0 | 75.3 | — |
|
97
DeepSeek V4 Flash
|
389.4 chars/s | 3.8B | 1 | — | — | — | — |
|
98
DeepSeek V4 Pro
|
208.2 chars/s | 4.2B | 1 | 0.9 | — | — | — |
|
99
Claude Sonnet 4.5
|
159.1 chars/s | 4.6B | 7–8 | 0.5 | 80.4 | 78.3 | 1397.0 |
|
100
Claude Haiku 4.5
|
334.0 chars/s | 5.0B | 1 | 0.2 | 72.8 | 71.4 | — |
How to read these statistics
Real usageUsage combines prompt and completion tokens generated by CodingFleet users during the selected period.
Credits and availabilityCredits are charged per request. One-credit models are unlimited on eligible Unlimited and Elite plans; legacy models are no longer selectable.
BenchmarksLiveBench measures broad and coding performance. WebDev Arena is an independent leaderboard for web-development output.