From a6eaeca02bbf7d80795384817b434ed376983221 Mon Sep 17 00:00:00 2001 From: "cursor-agent[bot]" Date: Mon, 13 Jul 2026 18:06:40 +0000 Subject: [PATCH] docs-sync: remove three serverless-deprecated models Sync with togethercomputer/mintlify-docs#1195, which deprecated zai-org/GLM-5.1, meta-llama/Meta-Llama-3-8B-Instruct-Lite, and Qwen/Qwen3-235B-A22B-Instruct-2507-tput from serverless on 2026-07-10. Also carry the follow-on recommendation changes: GLM-5.2 is now the recommended function-calling model, and Kimi K2.7 Code is the recommended coding-agent model. - references/models.md: drop the three deprecated rows from the full chat catalog; update the Coding Agents recommendation to Kimi K2.7 Code and the Function Calling recommendation to GLM-5.2 (matches docs/inference/recommended-models.mdx). - references/reasoning-models.md: drop GLM-5.1 from the full reasoning table and the reasoning={enabled} model list; rename the 'GLM-5.1 / GLM-5' best-practices section to 'GLM-5' and swap the remaining example model IDs from GLM-5.1 to GLM-5. - references/function-calling-patterns.md: drop GLM-5.1 and Qwen3-235B-A22B-Instruct-2507-tput from the supported-models paragraph. - references/structured-outputs.md: drop GLM-5.1 from the Top Models list. --- .../references/function-calling-patterns.md | 5 ++--- skills/together-chat-completions/references/models.md | 7 ++----- .../references/reasoning-models.md | 8 +++----- .../references/structured-outputs.md | 1 - 4 files changed, 7 insertions(+), 14 deletions(-) diff --git a/skills/together-chat-completions/references/function-calling-patterns.md b/skills/together-chat-completions/references/function-calling-patterns.md index 114c93e..df3eb6c 100644 --- a/skills/together-chat-completions/references/function-calling-patterns.md +++ b/skills/together-chat-completions/references/function-calling-patterns.md @@ -824,7 +824,6 @@ workflow. ## Supported Models openai/gpt-oss-120b, openai/gpt-oss-20b, moonshotai/Kimi-K2.6, -zai-org/GLM-5.1, zai-org/GLM-5, MiniMaxAI/MiniMax-M2.7, Qwen/Qwen3.5-397B-A17B, -Qwen/Qwen3.5-9B, Qwen/Qwen3.6-Plus, -Qwen/Qwen3-235B-A22B-Instruct-2507-tput, deepseek-ai/DeepSeek-V4-Pro, +zai-org/GLM-5, MiniMaxAI/MiniMax-M2.7, Qwen/Qwen3.5-397B-A17B, +Qwen/Qwen3.5-9B, Qwen/Qwen3.6-Plus, deepseek-ai/DeepSeek-V4-Pro, meta-llama/Llama-3.3-70B-Instruct-Turbo, Qwen/Qwen2.5-7B-Instruct-Turbo, google/gemma-4-31B-it diff --git a/skills/together-chat-completions/references/models.md b/skills/together-chat-completions/references/models.md index 39ea96f..3a5643f 100644 --- a/skills/together-chat-completions/references/models.md +++ b/skills/together-chat-completions/references/models.md @@ -6,10 +6,10 @@ |----------|-------|-----------|-------------| | Chat (best) | Kimi K2.6 | `moonshotai/Kimi-K2.6` | `MiniMaxAI/MiniMax-M2.7`, `openai/gpt-oss-120b` | | Reasoning | DeepSeek-V4-Pro | `deepseek-ai/DeepSeek-V4-Pro` | `moonshotai/Kimi-K2.6`, `Qwen/Qwen3.6-Plus` | -| Coding Agents | GLM-5.1 | `zai-org/GLM-5.1` | `moonshotai/Kimi-K2.6`, `deepseek-ai/DeepSeek-V4-Pro`, `MiniMaxAI/MiniMax-M2.7` | +| Coding Agents | Kimi K2.7 Code | `moonshotai/Kimi-K2.7-Code` | `zai-org/GLM-5.2`, `moonshotai/Kimi-K2.6`, `deepseek-ai/DeepSeek-V4-Pro` | | Small & Fast | GPT-OSS 20B | `openai/gpt-oss-20b` | `Qwen/Qwen2.5-7B-Instruct-Turbo`, `google/gemma-3n-E4B-it` | | Medium General | GPT-OSS 120B | `openai/gpt-oss-120b` | `zai-org/GLM-5` | -| Function Calling | GLM-5.1 | `zai-org/GLM-5.1` | `moonshotai/Kimi-K2.6`, `MiniMaxAI/MiniMax-M2.7` | +| Function Calling | GLM-5.2 | `zai-org/GLM-5.2` | `moonshotai/Kimi-K2.6`, `MiniMaxAI/MiniMax-M2.7` | | Vision | Qwen3.5 397B | `Qwen/Qwen3.5-397B-A17B` | `Qwen/Qwen3.5-9B`, `google/gemma-4-31B-it` | ## Full Chat Model Catalog @@ -21,16 +21,13 @@ | Qwen | Qwen3.5 397B A17B | `Qwen/Qwen3.5-397B-A17B` | 262,144 | BF16 | | Qwen | Qwen3.6 Plus | `Qwen/Qwen3.6-Plus` | 1,000,000 | - | | Qwen | Qwen3.5 9B | `Qwen/Qwen3.5-9B` | 262,144 | FP8 | -| Qwen | Qwen3 235B Instruct | `Qwen/Qwen3-235B-A22B-Instruct-2507-tput` | 262,144 | FP8 | | Moonshot | Kimi K2.6 | `moonshotai/Kimi-K2.6` | 262,144 | FP4 | | DeepSeek | DeepSeek-V4-Pro | `deepseek-ai/DeepSeek-V4-Pro` | 512,000 | FP4 | | NVIDIA | Nemotron 3 Ultra 550B A55B | `nvidia/nemotron-3-ultra-550b-a55b` | 512,300 | NVFP4 | | OpenAI | GPT-OSS 120B | `openai/gpt-oss-120b` | 128,000 | MXFP4 | | OpenAI | GPT-OSS 20B | `openai/gpt-oss-20b` | 128,000 | MXFP4 | -| Z.ai | GLM-5.1 | `zai-org/GLM-5.1` | 202,752 | FP4 | | Z.ai | GLM-5 | `zai-org/GLM-5` | 202,752 | FP4 | | Meta | Llama 3.3 70B Turbo | `meta-llama/Llama-3.3-70B-Instruct-Turbo` | 131,072 | FP8 | -| Meta | Llama 3 8B Lite | `meta-llama/Meta-Llama-3-8B-Instruct-Lite` | 8,192 | - | | Deep Cogito | Cogito v2.1 671B | `deepcogito/cogito-v2-1-671b` | 163,840 | - | | Google | Gemma 4 31B IT | `google/gemma-4-31B-it` | 262,144 | FP8 | | Google | Gemma 3N E4B | `google/gemma-3n-E4B-it` | 32,768 | FP8 | diff --git a/skills/together-chat-completions/references/reasoning-models.md b/skills/together-chat-completions/references/reasoning-models.md index 662723f..16632be 100644 --- a/skills/together-chat-completions/references/reasoning-models.md +++ b/skills/together-chat-completions/references/reasoning-models.md @@ -16,7 +16,6 @@ | Model | API String | Type | Context | Tool Calling | |-------|-----------|------|---------|--------------| | DeepSeek-V4-Pro | `deepseek-ai/DeepSeek-V4-Pro` | Hybrid (on by default) | 512K | Yes | -| GLM-5.1 | `zai-org/GLM-5.1` | Hybrid (on by default) | 200K | Yes | | GLM-5 | `zai-org/GLM-5` | Hybrid (on by default) | 200K | Yes | | GPT-OSS 120B | `openai/gpt-oss-120b` | Adjustable effort | 128K | No | | GPT-OSS 20B | `openai/gpt-oss-20b` | Adjustable effort | 128K | No | @@ -109,7 +108,6 @@ Hybrid models support `reasoning={"enabled": True/False}` to toggle reasoning on - `Qwen/Qwen3.6-Plus` (on by default) - `moonshotai/Kimi-K2.6` (on by default) - `nvidia/nemotron-3-ultra-550b-a55b` (on by default) -- `zai-org/GLM-5.1` (on by default) - `zai-org/GLM-5` (on by default) ### Python -- Enable Reasoning @@ -242,7 +240,7 @@ response = client.chat.completions.create( ### Separate reasoning field (most models) -Models like Kimi K2.6, GLM-5.1, DeepSeek-V4-Pro, GPT-OSS, and Qwen3.5 return reasoning in a dedicated +Models like Kimi K2.6, GLM-5, DeepSeek-V4-Pro, GPT-OSS, and Qwen3.5 return reasoning in a dedicated `reasoning` field on the response message or streaming delta. The field is symmetric: the model returns its chain of thought in `reasoning` (or `delta.reasoning` @@ -319,7 +317,7 @@ prompt tokens) always live somewhere on `response.usage`, but their **location v read both shapes defensively. - **Reasoning models** (for example `deepseek-ai/DeepSeek-V4-Pro`, `Qwen/Qwen3.6-Plus`, - `zai-org/GLM-5.1`) nest them OpenAI-style: + `zai-org/GLM-5`) nest them OpenAI-style: - `usage.completion_tokens_details.reasoning_tokens` - `usage.prompt_tokens_details.cached_tokens` - **Some non-reasoning models** (for example `meta-llama/Llama-3.3-70B-Instruct-Turbo`) return @@ -453,7 +451,7 @@ if (completion?.choices?.[0]?.message?.content) { - Supports both reasoning and non-reasoning modes - Excels at multi-turn tool calling with reasoning interleaved -### GLM-5.1 / GLM-5 +### GLM-5 - Thinking is enabled by default - Supports Preserved Thinking: set `"clear_thinking": false` in `chat_template_kwargs` - Preserved Thinking retains reasoning across turns for better agentic workflows diff --git a/skills/together-chat-completions/references/structured-outputs.md b/skills/together-chat-completions/references/structured-outputs.md index cf0ff32..7cc6979 100644 --- a/skills/together-chat-completions/references/structured-outputs.md +++ b/skills/together-chat-completions/references/structured-outputs.md @@ -509,7 +509,6 @@ console.log(`Title: ${result.title}`); - `openai/gpt-oss-120b` - `openai/gpt-oss-20b` - `moonshotai/Kimi-K2.6` -- `zai-org/GLM-5.1` - `zai-org/GLM-5` - `MiniMaxAI/MiniMax-M2.7` - `Qwen/Qwen3.5-397B-A17B`