InferenceBench
Europe
Providers
Models
Research
Partners
Data
Methodology
Search
/
PRICE
DeepSeek V4 Flash via OpenRouter: output price ↓ 33%
2m
PRICE
DeepSeek V4 Flash via OpenRouter: input price ↓ 33%
2m
PRICE
~moonshotai/kimi-latest input_per_mtok decreased from 1.39 to 1.34
3m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0086 to 0.0038
3m
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0086 to 0.0038
3m
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.012 to 0.0051
3m
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.012 to 0.0051
3m
PRICE
deepseek/deepseek-v4-flash cached_input_per_mtok decreased from 0.0084 to 0.0056
3m
PRICE
deepseek/deepseek-v4-flash output_per_mtok decreased from 0.084 to 0.056
3m
PRICE
deepseek/deepseek-v4-flash input_per_mtok decreased from 0.042 to 0.028
3m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.013 to 0.0086
34m
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.013 to 0.0086
34m
CAPABILITY
xiaomi/mimo-v2.6-flash now declares support for top_logprobs
34m
CAPABILITY
xiaomi/mimo-v2.6-flash now declares support for logprobs
34m
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.017 to 0.012
34m
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.017 to 0.012
34m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok increased from 0.0058 to 0.013
1h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok increased from 0.0058 to 0.013
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for stop
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for frequency_penalty
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for stop
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for frequency_penalty
1h
CAPABILITY
openai/gpt-5-image no longer declares support for stop
1h
CAPABILITY
openai/gpt-5-image no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5-image no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5-image no longer declares support for frequency_penalty
1h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok increased from 0.0077 to 0.017
1h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok increased from 0.0077 to 0.017
1h
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0086 to 0.0058
2h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0086 to 0.0058
2h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.012 to 0.0077
2h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.012 to 0.0077
2h
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.013 to 0.0086
2h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.013 to 0.0086
2h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.017 to 0.012
2h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.017 to 0.012
2h
PRICE
DeepSeek V4 Flash via OpenRouter: output price ↓ 33%
2m
PRICE
DeepSeek V4 Flash via OpenRouter: input price ↓ 33%
2m
PRICE
~moonshotai/kimi-latest input_per_mtok decreased from 1.39 to 1.34
3m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0086 to 0.0038
3m
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0086 to 0.0038
3m
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.012 to 0.0051
3m
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.012 to 0.0051
3m
PRICE
deepseek/deepseek-v4-flash cached_input_per_mtok decreased from 0.0084 to 0.0056
3m
PRICE
deepseek/deepseek-v4-flash output_per_mtok decreased from 0.084 to 0.056
3m
PRICE
deepseek/deepseek-v4-flash input_per_mtok decreased from 0.042 to 0.028
3m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.013 to 0.0086
34m
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.013 to 0.0086
34m
CAPABILITY
xiaomi/mimo-v2.6-flash now declares support for top_logprobs
34m
CAPABILITY
xiaomi/mimo-v2.6-flash now declares support for logprobs
34m
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.017 to 0.012
34m
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.017 to 0.012
34m
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok increased from 0.0058 to 0.013
1h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok increased from 0.0058 to 0.013
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for stop
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5.4-image-2 no longer declares support for frequency_penalty
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for stop
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5-image-mini no longer declares support for frequency_penalty
1h
CAPABILITY
openai/gpt-5-image no longer declares support for stop
1h
CAPABILITY
openai/gpt-5-image no longer declares support for presence_penalty
1h
CAPABILITY
openai/gpt-5-image no longer declares support for logit_bias
1h
CAPABILITY
openai/gpt-5-image no longer declares support for frequency_penalty
1h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok increased from 0.0077 to 0.017
1h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok increased from 0.0077 to 0.017
1h
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.0086 to 0.0058
2h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.0086 to 0.0058
2h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.012 to 0.0077
2h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.012 to 0.0077
2h
PRICE
~deepseek/deepseek-v4-flash-latest cached_input_per_mtok decreased from 0.013 to 0.0086
2h
PRICE
~deepseek/deepseek-v4-flash-latest input_per_mtok decreased from 0.013 to 0.0086
2h
PRICE
deepseek/deepseek-v4-flash-0731 cached_input_per_mtok decreased from 0.017 to 0.012
2h
PRICE
deepseek/deepseek-v4-flash-0731 input_per_mtok decreased from 0.017 to 0.012
2h
Qwen3-235B
on
OpenRouter
#2 of 3 providers for Qwen3-235B · #4 of 13 models on OpenRouter · US · EU
Compare with other providers
Score
89.5
Rel.
100.0%
Tools
100.0%
JSON
100.0%
TTFT
766 ms
P95
3027 ms
Tok/s
59
Price
$0.087 / $0.35
Ctx
128k
217 results · last 275m ago
Qwen3-235B on other providers
Provider
Score
TTFT
Tok/s
$ Out
Scaleway
91.8
162 ms
81
$2.58
OpenRouter
THIS PAGE
89.5
766 ms
59
$0.35
Cortecs
86.3
937 ms
52
$0.47
Other models on OpenRouter
Model
Score
TTFT
$ Out
GPT-OSS 120B
93.3
474 ms
$0.17
GPT-OSS 20B
92.3
656 ms
$0.09
Llama 3.3 70B
89.4
576 ms
$0.32
Qwen3-235B
THIS PAGE
86.2
766 ms
$0.35
DeepSeek V4 Flash
85.9
1059 ms
$0.056
Mistral Large 3
81.5
406 ms
$1.50
Mistral Medium 3.5
80.3
261 ms
$7.50
DeepSeek V4 Pro
80.2
1086 ms
$0.42
GLM-5.3
79.8
412 ms
$4.40
MiniMax M3
77.8
929 ms
$1.20
GLM-5.2
73.8
480 ms
$3.99
Kimi K3
72.0
674 ms
$13.50
GLM-4.7
59.9
2259 ms
$2.20