Skip to content

fireworks-ai: add missing low reasoning effort tier to glm-5p3 and glm-5p3-flash #6237

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
rekram1-node merged 1 commit into from
Sep 4, 2026
+14 −9
Merged

fireworks-ai: add missing low reasoning effort tier to glm-5p3 and glm-5p3-flash #6237

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
11 changes: 7 additions & 4 deletions providers/fireworks-ai/models/accounts/fireworks/models/glm-5p3-flash.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,14 +1,17 @@
# Pricing: https://fireworks.ai/models/fireworks/glm-5p3-flash (accessed 2026-09-02)
# Released on Fireworks Serverless API 2026-08-26.
# Thinking-only model: reasoning cannot be disabled. Use the same two effective
# tiers as Fireworks GLM 5.3: low/medium collapse to high; max/xhigh select max.
# https://docs.fireworks.ai/api-reference/post-chatcompletions (accessed 2026-08-29)
# Thinking-only model: reasoning cannot be disabled. Three distinct effort
# tiers like Fireworks GLM 5.3 (NOT the two-tier GLM 5.2 collapse): verified
# empirically 2026-09-04 — same prompt yields ~9/17/559 reasoning tokens for
# low/high/max. max/xhigh select max.
# https://docs.fireworks.ai/api-reference/post-chatcompletions
base_model = "zhipuai/glm-5.3-flash"
name = "GLM 5.3 Flash"
last_updated = "2026-09-04"

[[reasoning_options]]
type = "effort"
values = ["high", "max"]
values = ["low", "high", "max"]

[interleaved]
field = "reasoning_content"
Expand Down
12 changes: 7 additions & 5 deletions providers/fireworks-ai/models/accounts/fireworks/models/glm-5p3.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,16 +1,18 @@
# Pricing: https://docs.fireworks.ai/serverless/pricing (accessed 2026-08-29)
# Released on Fireworks Serverless API 2026-08-28.
# Thinking-only model: reasoning cannot be disabled (reasoning_effort="none"
# and thinking.type="disabled" both return 400). Two effective tiers like
# GLM 5.2: low/medium collapse to high, max/xhigh select max.
# https://docs.fireworks.ai/api-reference/post-chatcompletions (accessed 2026-08-29)
# and thinking.type="disabled" both return 400). Unlike GLM 5.2 (two tiers,
# low/medium collapse to high), GLM 5.3 has three distinct effort tiers:
# verified empirically 2026-09-04 — same prompt yields ~10/25/198 reasoning
# tokens for low/high/max. max/xhigh select max.
# https://docs.fireworks.ai/api-reference/post-chatcompletions
base_model = "zhipuai/glm-5.3"
name = "GLM 5.3"
last_updated = "2026-08-28"
last_updated = "2026-09-04"

[[reasoning_options]]
type = "effort"
values = ["high", "max"]
values = ["low", "high", "max"]

[interleaved]
field = "reasoning_content"
Expand Down
Loading