Skip to content

Add new models (2025-10-17) #19

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’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Yuki-Imajuku merged 1 commit into from
Oct 27, 2025
Merged

Add new models (2025-10-17) #19

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
17 changes: 17 additions & 0 deletions llm_configs/claude-4.5-haiku.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,17 @@
{
"model_name": "anthropic/claude-haiku-4.5",
"provider": "openrouter",
"settings": {
"temperature": 1.0,
"max_tokens": 64000,
"extra_body": {
"reasoning": {
"enabled": true,
"max_tokens": 32000
},
"usage": {
"include": true
}
}
}
}
2 changes: 1 addition & 1 deletion llm_configs/deepseek-v3.1-terminus.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
"provider": {
"allow_fallbacks": false,
"data_collection": "deny",
"order": ["siliconflow/fp8"],
"order": ["deepinfra/fp4"],
"require_parameters": true
},
"reasoning": {
Expand Down
3 changes: 1 addition & 2 deletions llm_configs/glm-4.6.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,11 +5,10 @@
"temperature": 1.0,
"top_p": 0.95,
"extra_body": {
"top_k": 40,
"provider": {
"allow_fallbacks": false,
"data_collection": "deny",
"order": ["parasail/fp8"],
"order": ["z-ai"],
"require_parameters": true
},
"reasoning": {
Expand Down
15 changes: 15 additions & 0 deletions llm_configs/qwen3-coder-plus.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
{
"model_name": "qwen/qwen3-coder-plus",
"provider": "openrouter",
"settings": {
"temperature": 0.7,
"max_tokens": 65536,
"top_p": 0.8,
"extra_body": {
"top_k": 20,
"usage": {
"include": true
}
}
}
}
26 changes: 25 additions & 1 deletion src/ale_bench_eval/calc_cost.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,11 +36,21 @@
),
output_mtok=TieredPrices(base=Decimal(15), tiers=[Tier(start=200000, price=Decimal(225) / Decimal(10))]),
),
"claude-haiku-4.5": ModelPrice(
input_mtok=Decimal(1),
cache_write_mtok=Decimal(125) / Decimal(100),
cache_read_mtok=Decimal(1) / Decimal(10),
output_mtok=Decimal(5),
),
"deepseek-v3.1": ModelPrice(input_mtok=Decimal(56) / Decimal(100), output_mtok=Decimal(168) / Decimal(100)),
"deepseek-v3.1-terminus": ModelPrice(input_mtok=Decimal(27) / Decimal(100), output_mtok=Decimal(1)),
"deepseek-r1-0528": ModelPrice(input_mtok=Decimal(79) / Decimal(100), output_mtok=Decimal(4)),
"glm-4.5": ModelPrice(input_mtok=Decimal(59) / Decimal(100), output_mtok=Decimal(21) / Decimal(10)),
"glm-4.6": ModelPrice(input_mtok=Decimal(6) / Decimal(10), output_mtok=Decimal(21) / Decimal(10)),
"glm-4.6": ModelPrice(
input_mtok=Decimal(6) / Decimal(10),
Copy link

Copilot AI Oct 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[nitpick] The output_mtok price for glm-4.6 changed from 2.1 to 2.2. Consider documenting this pricing update in the commit message or changelog to clarify whether this is a correction or reflects an actual price change by the provider.

Suggested change
input_mtok=Decimal(6) / Decimal(10),
input_mtok=Decimal(6) / Decimal(10),
# Updated output_mtok price for glm-4.6 from 2.1 to 2.2 as per provider's latest pricing (June 2024)

Copilot uses AI. Check for mistakes.
output_mtok=Decimal(22) / Decimal(10),
cache_read_mtok=Decimal(11) / Decimal(100),
),
"gpt-oss-120b": ModelPrice(input_mtok=Decimal(1) / Decimal(10), output_mtok=Decimal(5) / Decimal(10)),
"gpt-oss-20b": ModelPrice(input_mtok=Decimal(5) / Decimal(100), output_mtok=Decimal(2) / Decimal(10)),
"grok-code-fast-1": ModelPrice(input_mtok=Decimal(2) / Decimal(10), output_mtok=Decimal(15) / Decimal(10)),
Expand All @@ -61,6 +71,20 @@
input_mtok=Decimal(3) / Decimal(10), output_mtok=Decimal(29) / Decimal(10)
),
"qwen3-coder": ModelPrice(input_mtok=Decimal(29) / Decimal(100), output_mtok=Decimal(12) / Decimal(10)),
"qwen3-coder-plus": ModelPrice(
input_mtok=TieredPrices(
base=Decimal(1),
tiers=[Tier(start=32000, price=Decimal(18) / Decimal(10))],
),
output_mtok=TieredPrices(
base=Decimal(5),
tiers=[Tier(start=32000, price=Decimal(9))],
),
cache_read_mtok=TieredPrices(
base=Decimal(1) / Decimal(10),
tiers=[Tier(start=32000, price=Decimal(18) / Decimal(100))],
),
),
"qwen3-max": ModelPrice(
input_mtok=TieredPrices(base=Decimal(12) / Decimal(10), tiers=[Tier(start=128000, price=Decimal(3))]),
cache_read_mtok=TieredPrices(
Expand Down