All General Coding Reasoning Multimodal Safety Specialized Agentic Language Long Context Speed Text-to-Image Image Editing Text-to-Video Image-to-Video
Generation Tokens/sec incl. Reasoning (RightRank Probe)113 Inter-Token Latency Latency Output Tokens/sec163 Output Tokens/sec (RightRank Probe)165 TTFT163 TTFT (RightRank Probe)180 Time to First Answer Token (RightRank Probe)100 Tool Call TTFT (RightRank Probe)169 Tool Call Total Time (RightRank Probe)169 Total Time (RightRank Probe)181

Total Time (RightRank Probe)

Speed

Time to the full answer of a prompt that asks for about 500 tokens: median of 5 streamed runs through OpenRouter, pinned to the model's first-party provider, from the us-east-1 region.

Lower is better

Rank Model Provider Time
1 Inception: Mercury 2.5 Inception
1.4 s p90 1.8 s · 5 runs
2 Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) Google
1.6 s p90 1.8 s · 5 runs
3 Google: Gemini 2.5 Flash LiteRetiring Google
1.8 s p90 2.0 s · 5 runs
4 Inception: Mercury 2 Inception
1.8 s p90 2.0 s · 5 runs
5 Amazon: Nova Micro 1.0 Amazon
2.0 s p90 2.4 s · 5 runs
6 Mistral: Ministral 3 3B 2512 Mistral AI
2.3 s p90 2.5 s · 5 runs
7 MiniMax: MiniMax M2-her MiniMax
2.6 s p90 7.0 s · 5 runs
8 Google: Gemini 3.1 Flash Lite Preview Google
2.7 s p90 3.0 s · 5 runs
9 Google: Gemini 3.1 Flash Lite Google
2.8 s p90 3.0 s · 5 runs
10 Mistral: Voxtral Small 24B 2507 Mistral AI
2.8 s p90 3.4 s · 5 runs
11 Mistral: Mistral Medium 3.1 Mistral AI
2.9 s p90 3.2 s · 5 runs
12 Qwen: Qwen3 VL 30B A3B Instruct Alibaba
3.0 s p90 3.2 s · 5 runs
13 OpenAI: GPT-5.4 Nano OpenAI
3.1 s p90 3.6 s · 5 runs
14 OpenAI: GPT-3.5 Turbo 16k OpenAI
3.1 s p90 3.5 s · 5 runs
15 OpenAI: o3 Mini OpenAI
3.2 s p90 3.6 s · 5 runs
16 OpenAI: GPT-3.5 Turbo OpenAI
3.2 s p90 3.5 s · 5 runs
17 Amazon: Nova Pro 1.0 Amazon
3.2 s p90 4.5 s · 5 runs
18 Google: Gemini 3.5 Flash Lite Google
3.3 s p90 4.3 s · 5 runs
19 Mistral: Mistral Medium 3.5 Mistral AI
3.3 s p90 3.5 s · 5 runs
20 Mistral: Devstral 2 2512 Mistral AI
3.4 s p90 3.8 s · 5 runs
21 Google: Gemini 2.5 FlashRetiring Google
3.4 s p90 3.6 s · 5 runs
22 Google: Gemini 3.8 Flash Google
3.4 s p90 3.5 s · 5 runs
23 OpenAI: GPT-4.1 Nano OpenAI
3.4 s p90 4.8 s · 5 runs
24 Amazon: Nova Lite 1.0 Amazon
3.4 s p90 4.0 s · 5 runs
25 Amazon: Nova 2 Lite Amazon
3.5 s p90 3.8 s · 5 runs
26 Mistral: Mistral Medium 3 Mistral AI
3.5 s p90 3.8 s · 5 runs
27 Google: Gemini 3.7 Flash Google
3.5 s p90 4.2 s · 5 runs
28 OpenAI: o4 Mini OpenAI
3.6 s p90 4.5 s · 5 runs
45 ↳OpenAI: o4 Mini High OpenAI
4.4 s p90 7.3 s · 5 runs
29 Qwen: Qwen3 Coder Flash Alibaba
3.7 s p90 4.2 s · 5 runs
30 Qwen: Qwen3 Coder Next Alibaba
3.7 s p90 3.7 s · 5 runs
31 Perplexity: Sonar Perplexity
3.7 s p90 4.5 s · 5 runs
32 Qwen: Qwen3.5-Flash Alibaba
3.7 s p90 4.2 s · 5 runs
33 Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview) Google
3.7 s p90 4.0 s · 5 runs
34 Google: Nano Banana 2 (Gemini 3.1 Flash Image) Google
3.7 s p90 4.3 s · 5 runs
35 Google: Gemini 3 Flash Preview Google
3.8 s p90 3.9 s · 5 runs
36 Google: Gemini 3.5 Flash Google
3.8 s p90 4.3 s · 5 runs
37 OpenAI: GPT-4.1 Mini OpenAI
3.8 s p90 5.0 s · 5 runs
38 OpenAI: GPT-6 Luna OpenAI
3.9 s p90 4.4 s · 5 runs
39 Qwen: Qwen3 Coder 30B A3B Instruct Alibaba
3.9 s p90 4.1 s · 5 runs
40 OpenAI: GPT-4o-mini (2024-07-18) OpenAI
4.2 s p90 4.7 s · 5 runs
41 Google: Gemini 3.6 Flash Google
4.2 s p90 4.5 s · 5 runs
42 OpenAI: GPT-4o (2024-05-13) OpenAI
4.3 s p90 4.8 s · 5 runs
43 Qwen: Qwen3 VL 8B InstructHistorical Alibaba
4.3 s p90 4.6 s · 5 runs
44 OpenAI: GPT-4o-mini OpenAI
4.4 s p90 4.8 s · 5 runs
46 Qwen: Qwen3 30B A3B Instruct 2507 Alibaba
4.5 s p90 4.7 s · 5 runs
47 OpenAI: GPT-5.6 Luna OpenAI
4.5 s p90 4.7 s · 5 runs
48 OpenAI: GPT-4o OpenAI
4.5 s p90 5.2 s · 5 runs
49 Mistral: Codestral 2508 Mistral AI
4.6 s p90 4.8 s · 5 runs
50 Qwen: Qwen3 Next 80B A3B InstructHistorical Alibaba
4.6 s p90 4.8 s · 5 runs

Access this data

JSON GET /api/v1 
[
  {
    "rank": 1,
    "model": "Inception: Mercury 2.5",
    "provider": "Inception",
    "score": "1394.8"
  },
  {
    "rank": 2,
    "model": "Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)",
    "provider": "Google",
    "score": "1630.8"
  },
  {
    "rank": 3,
    "model": "Google: Gemini 2.5 Flash Lite",
    "provider": "Google",
    "score": "1810.4"
  }
]
Curl 
$ curl -s https://rightrank.ai/api/v1/rankings/total-time-rightrank-probe.json | jq '.[:3]'
CLI brew install rightrank 
$ rightrank top total-time-rightrank-probe --limit 5
$ rightrank compare claude-opus-4.6 gpt-5.4-pro --category total-time-rightrank-probe
$ rightrank value total-time-rightrank-probe  # best performance per dollar
Claude Code Skill Skill 
# In Claude Code, use the RightRank skill:
/rightrank best model for total-time-rightrank-probe
/rightrank compare top 5 by value
/rightrank cheapest model scoring above 80 on total-time-rightrank-probe