Skip to content

Add Tempr's Cohere, AI21, Inception, Morph and Sarvam models #8418

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
chrissalomon wants to merge 1 commit into
base: dev
Choose a base branch
from
Open

Add Tempr's Cohere, AI21, Inception, Morph and Sarvam models #8418

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions providers/tempr/models/ai21/jamba-large.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "ai21/jamba-large"

[cost]
input = 2
output = 8
5 changes: 5 additions & 0 deletions providers/tempr/models/ai21/jamba-mini.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "ai21/jamba-mini"

[cost]
input = 0.2
output = 0.4
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-a-03-2025.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-a-03-2025"

[cost]
input = 2.5
output = 10
15 changes: 15 additions & 0 deletions providers/tempr/models/cohere/command-a-plus-05-2026.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off)
# Budget: reasoning.max_tokens (integer reasoning tokens)
# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens
# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget)
base_model = "cohere/command-a-plus-05-2026"

reasoning_options = [
{ type = "toggle" },
{ type = "budget_tokens", min = 1 },
]

[cost]
input = 2.5
output = 10
15 changes: 15 additions & 0 deletions providers/tempr/models/cohere/command-a-reasoning-08-2025.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off)
# Budget: reasoning.max_tokens (integer reasoning tokens)
# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens
# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget)
base_model = "cohere/command-a-reasoning-08-2025"

reasoning_options = [
{ type = "toggle" },
{ type = "budget_tokens", min = 1 },
]

[cost]
input = 2.5
output = 10
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-a-translate-08-2025.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-a-translate-08-2025"

[cost]
input = 2.5
output = 10
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-a-vision-07-2025.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-a-vision-07-2025"

[cost]
input = 2.5
output = 10
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-r-08-2024.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-r-08-2024"

[cost]
input = 0.15
output = 0.6
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-r-plus-08-2024.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-r-plus-08-2024"

[cost]
input = 2.5
output = 10
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-r7b-12-2024.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-r7b-12-2024"

[cost]
input = 0.0375
output = 0.15
5 changes: 5 additions & 0 deletions providers/tempr/models/cohere/command-r7b-arabic-02-2025.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "cohere/command-r7b-arabic-02-2025"

[cost]
input = 0.0375
output = 0.15
14 changes: 14 additions & 0 deletions providers/tempr/models/cohere/north-mini-code-1-0.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# Off: reasoning.effort = "none" (or reasoning.enabled = false)
# Effort: reasoning.effort = none|high (alias: top-level reasoning_effort)
# /v1/messages: thinking.type = enabled|disabled; output_config.effort = high
# /v1/responses: reasoning.effort = none|high
base_model = "cohere/north-mini-code-1-0"

reasoning_options = [
{ type = "effort", values = ["none", "high"] },
]

[cost]
input = 0
output = 0
14 changes: 14 additions & 0 deletions providers/tempr/models/inception/mercury-2.5.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort)
# /v1/messages: output_config.effort = low|medium|high
# /v1/responses: reasoning.effort = low|medium|high
base_model = "inception/mercury-2.5"

reasoning_options = [
{ type = "effort", values = ["low", "medium", "high"] },
]

[cost]
input = 0.04
output = 0.15
cache_read = 0.004
14 changes: 14 additions & 0 deletions providers/tempr/models/inception/mercury-2.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort)
# /v1/messages: output_config.effort = low|medium|high
# /v1/responses: reasoning.effort = low|medium|high
base_model = "inception/mercury-2"

reasoning_options = [
{ type = "effort", values = ["low", "medium", "high"] },
]

[cost]
input = 0.25
output = 0.75
cache_read = 0.025
5 changes: 5 additions & 0 deletions providers/tempr/models/morph/morph-v3-fast.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "morph/morph-v3-fast"

[cost]
input = 0.8
output = 1.2
5 changes: 5 additions & 0 deletions providers/tempr/models/morph/morph-v3-large.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
base_model = "morph/morph-v3-large"

[cost]
input = 0.9
output = 1.9
11 changes: 11 additions & 0 deletions providers/tempr/models/sarvam/sarvam-105b.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning
# No off switch on Tempr: Sarvam switches thinking off with a JSON null effort, which Tempr
# doesn't send, so reasoning.effort = "none" maps to the lowest level below.
# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort)
# /v1/messages: output_config.effort = low|medium|high
# /v1/responses: reasoning.effort = low|medium|high
base_model = "sarvam/sarvam-105b"

reasoning_options = [
{ type = "effort", values = ["low", "medium", "high"] },
]
Loading