summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--config.example.yaml2
-rw-r--r--models.yaml63
-rw-r--r--src/client/model.rs10
-rw-r--r--src/client/openai.rs3
4 files changed, 12 insertions, 66 deletions
diff --git a/config.example.yaml b/config.example.yaml
index 43cec93..2a9168c 100644
--- a/config.example.yaml
+++ b/config.example.yaml
@@ -95,7 +95,6 @@ clients:
# max_input_tokens: 100000
# supports_vision: true
# supports_function_calling: true
- # supports_reasoning: true
# - name: xxxx # Embedding model
# type: embedding
# default_chunk_size: 1500
@@ -128,7 +127,6 @@ clients:
models:
- name: deepseek-r1
max_input_tokens: 131072
- supports_reasoning: true
- name: llama3.1
max_input_tokens: 128000
supports_function_calling: true
diff --git a/models.yaml b/models.yaml
index 5c8511c..547081e 100644
--- a/models.yaml
+++ b/models.yaml
@@ -38,7 +38,6 @@
output_price: 4.4
supports_vision: true
supports_function_calling: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -50,7 +49,6 @@
output_price: 60
supports_vision: true
supports_function_calling: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -61,7 +59,6 @@
max_output_tokens: 32768
input_price: 15
output_price: 60
- supports_reasoning: true
no_system_message: true
patch:
body:
@@ -72,7 +69,6 @@
max_output_tokens: 65536
input_price: 3
output_price: 12
- supports_reasoning: true
no_system_message: true
patch:
body:
@@ -130,7 +126,6 @@
input_price: 0
output_price: 0
supports_vision: true
- supports_reasoning: true
- name: gemini-2.0-pro-exp
max_input_tokens: 2097152
max_output_tokens: 8192
@@ -405,12 +400,10 @@
max_input_tokens: 127000
input_price: 2
output_price: 8
- supports_reasoning: true
- name: sonar-reasoning
max_input_tokens: 127000
input_price: 1
output_price: 5
- supports_reasoning: true
# Links:
# - https://console.groq.com/docs/models
@@ -441,7 +434,6 @@
max_input_tokens: 131072
input_price: 0
output_price: 0
- supports_reasoning: true
# Links:
# - https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models
@@ -473,7 +465,6 @@
max_input_tokens: 32760
max_output_tokens: 8192
supports_vision: true
- supports_reasoning: true
- name: gemini-2.0-pro-exp-02-05
max_input_tokens: 2097152
max_output_tokens: 8192
@@ -759,7 +750,6 @@
require_max_tokens: true
input_price: 0
output_price: 0
- supports_reasoning: true
- name: '@cf/baai/bge-large-en-v1.5'
type: embedding
input_price: 0
@@ -857,14 +847,12 @@
max_input_tokens: 16384
max_output_tokens: 16384
supports_vision: true
- supports_reasoning: true
- name: qwq-32b-preview
max_input_tokens: 30720
max_output_tokens: 16384
input_price: 0.49
output_price: 0.98
supports_function_calling: true
- supports_reasoning: true
- name: qwen-vl-max-latest
max_input_tokens: 30720
max_output_tokens: 2048
@@ -1021,7 +1009,6 @@
max_output_tokens: 8192
input_price: 0.55
output_price: 2.19
- supports_reasoning: true
# Links:
# - https://open.bigmodel.cn/pricing
@@ -1066,7 +1053,6 @@
max_input_tokens: 16384
input_price: 1.4
output_price: 1.4
- supports_reasoning: true
- name: embedding-3
type: embedding
max_input_tokens: 8192
@@ -1155,7 +1141,6 @@
output_price: 4.4
supports_vision: true
supports_function_calling: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -1167,7 +1152,6 @@
output_price: 60
supports_vision: true
supports_function_calling: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -1177,7 +1161,6 @@
max_input_tokens: 128000
input_price: 15
output_price: 60
- supports_reasoning: true
no_system_message: true
patch:
body:
@@ -1187,7 +1170,6 @@
max_input_tokens: 128000
input_price: 3
output_price: 12
- supports_reasoning: true
no_system_message: true
patch:
body:
@@ -1358,17 +1340,23 @@
max_input_tokens: 163840
input_price: 0.55
output_price: 2.19
- supports_reasoning: true
+ patch:
+ body:
+ include_reasoning: true
- name: deepseek/deepseek-r1-distill-llama-70b
max_input_tokens: 131072
input_price: 0.23
output_price: 0.69
- supports_reasoning: true
+ patch:
+ body:
+ include_reasoning: true
- name: deepseek/deepseek-r1-distill-qwen-32b
max_input_tokens: 131072
input_price: 0.12
output_price: 0.18
- supports_reasoning: true
+ patch:
+ body:
+ include_reasoning: true
- name: qwen/qwen-max
max_input_tokens: 32768
max_output_tokens: 8192
@@ -1404,13 +1392,11 @@
max_input_tokens: 32768
input_price: 0.15
output_price: 0.6
- supports_reasoning: true
- name: qwen/qvq-72b-preview
max_input_tokens: 128000
input_price: 0.25
output_price: 0.5
supports_vision: true
- supports_reasoning: true
- name: x-ai/grok-2-1212
max_input_tokens: 131072
input_price: 2
@@ -1453,7 +1439,9 @@
max_input_tokens: 127000
input_price: 1
output_price: 5
- supports_reasoning: true
+ patch:
+ body:
+ include_reasoning: true
- name: perplexity/sonar
max_input_tokens: 127000
input_price: 1
@@ -1478,7 +1466,6 @@
max_input_tokens: 200000
supports_function_calling: true
supports_vision: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -1488,7 +1475,6 @@
max_input_tokens: 200000
supports_function_calling: true
supports_vision: true
- supports_reasoning: true
system_prompt_prefix: Formatting re-enabled
patch:
body:
@@ -1496,7 +1482,6 @@
top_p: null
- name: o1-preview
max_input_tokens: 128000
- supports_reasoning: true
no_stream: true
no_system_message: true
patch:
@@ -1505,7 +1490,6 @@
top_p: null
- name: o1-mini
max_input_tokens: 128000
- supports_reasoning: true
no_stream: true
no_system_message: true
patch:
@@ -1569,7 +1553,6 @@
supports_function_calling: true
- name: deepseek-r1
max_input_tokens: 163840
- supports_reasoning: true
- name: phi-4
max_input_tokens: 16384
- name: phi-3.5-moe-instruct
@@ -1625,12 +1608,10 @@
input_price: 0.25
output_price: 0.50
supports_vision: true
- supports_reasoning: true
- name: Qwen/QwQ-32B-Preview
max_input_tokens: 32768
input_price: 0.12
output_price: 0.18
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-V3
max_input_tokens: 32768
input_price: 0.85
@@ -1639,17 +1620,14 @@
max_input_tokens: 16000
input_price: 0.85
output_price: 2.5
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
max_input_tokens: 131072
input_price: 0.23
output_price: 0.69
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
max_input_tokens: 131072
input_price: 0.12
output_price: 0.18
- supports_reasoning: true
- name: mistralai/Mistral-Small-24B-Instruct-2501
max_input_tokens: 32768
input_price: 0.07
@@ -1735,7 +1713,6 @@
max_input_tokens: 32768
input_price: 0.9
output_price: 0.9
- supports_reasoning: true
- name: accounts/fireworks/models/qwen2-vl-72b-instruct
max_input_tokens: 32768
input_price: 0.9
@@ -1749,17 +1726,14 @@
max_input_tokens: 160000
input_price: 8
output_price: 8
- supports_reasoning: true
- name: accounts/fireworks/models/deepseek-r1-distill-llama-70b
max_input_tokens: 131072
input_price: 0.9
output_price: 0.9
- supports_reasoning: true
- name: accounts/fireworks/models/deepseek-r1-distill-qwen-32b
max_input_tokens: 131072
input_price: 0.9
output_price: 0.9
- supports_reasoning: true
- name: accounts/fireworks/models/mistral-small-24b-instruct-2501
max_input_tokens: 32768
input_price: 0.9
@@ -1837,12 +1811,10 @@
input_price: 1.386
output_price: 1.386
supports_vision: true
- supports_reasoning: true
- name: Qwen/QwQ-32B-Preview
max_input_tokens: 32768
input_price: 0.176
output_price: 0.176
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-V3
max_input_tokens: 65536
input_price: 0.28
@@ -1851,17 +1823,14 @@
max_input_tokens: 65536
input_price: 2.24
output_price: 2.24
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
max_input_tokens: 32768
input_price: 0.578
output_price: 0.578
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
max_input_tokens: 32768
input_price: 0.176
output_price: 0.176
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-V2.5
max_input_tokens: 32768
input_price: 0.7
@@ -1946,7 +1915,6 @@
max_input_tokens: 32768
input_price: 1.2
output_price: 1.2
- supports_reasoning: true
- name: Qwen/Qwen2-VL-72B-Instruct
max_input_tokens: 32768
input_price: 1.2
@@ -1960,12 +1928,10 @@
max_input_tokens: 163840
input_price: 7
output_price: 7
- supports_reasoning: true
- name: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
max_input_tokens: 131072
input_price: 2
output_price: 2
- supports_reasoning: true
- name: mistralai/Mistral-Small-24B-Instruct-2501
max_input_tokens: 32768
input_price: 0.8
@@ -2020,7 +1986,6 @@
max_input_tokens: 32768
input_price: 0.2
output_price: 0.2
- supports_reasoning: true
- name: Qwen/Qwen2-VL-72B-Instruct
max_input_tokens: 32768
input_price: 0.4
@@ -2034,7 +1999,6 @@
max_input_tokens: 163840
input_price: 2
output_price: 2
- supports_reasoning: true
- name: mistralai/Pixtral-12B-2409
max_input_tokens: 128000
input_price: 0.1
@@ -2080,17 +2044,14 @@
max_input_tokens: 64000
input_price: 4
output_price: 4
- supports_reasoning: true
- name: deepseek/deepseek-r1-distill-llama-70b
max_input_tokens: 32000
input_price: 0.8
output_price: 0.8
- supports_reasoning: true
- name: deepseek/deepseek-r1-distill-qwen-32b
max_input_tokens: 128000
input_price: 0.3
output_price: 0.3
- supports_reasoning: true
- name: mistralai/mistral-nemo
max_input_tokens: 131072
input_price: 0.17
diff --git a/src/client/model.rs b/src/client/model.rs
index 82e2252..20996b8 100644
--- a/src/client/model.rs
+++ b/src/client/model.rs
@@ -136,7 +136,6 @@ impl Model {
output_price,
supports_vision,
supports_function_calling,
- supports_reasoning,
..
} = &self.data;
let max_input_tokens = stringify_option_value(max_input_tokens);
@@ -150,9 +149,6 @@ impl Model {
if *supports_function_calling {
capabilities.push('⚒');
};
- if *supports_reasoning {
- capabilities.push('💭');
- }
let capabilities: String = capabilities
.into_iter()
.map(|v| format!("{v} "))
@@ -191,10 +187,6 @@ impl Model {
self.data.max_output_tokens
}
- pub fn supports_reasoning(&self) -> bool {
- self.data.supports_reasoning
- }
-
pub fn no_stream(&self) -> bool {
self.data.no_stream
}
@@ -327,8 +319,6 @@ pub struct ModelData {
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
pub supports_function_calling: bool,
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
- supports_reasoning: bool,
- #[serde(default, skip_serializing_if = "std::ops::Not::not")]
no_stream: bool,
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
no_system_message: bool,
diff --git a/src/client/openai.rs b/src/client/openai.rs
index 975afd7..6d57d42 100644
--- a/src/client/openai.rs
+++ b/src/client/openai.rs
@@ -301,9 +301,6 @@ pub fn openai_build_chat_completions_body(data: ChatCompletionsData, model: &Mod
if let Some(v) = model.max_tokens_param() {
body["max_tokens"] = v.into();
}
- if model.client_name().starts_with("openrouter") && model.supports_reasoning() {
- body["include_reasoning"] = true.into();
- }
if let Some(v) = temperature {
body["temperature"] = v.into();
}
generated by cgit v1.2.3 (git 2.39.1) at 2026-10-03 00:50:39 +0000