From 98cc1ffc61bb09dd70d4ce0072163fc7783f3901 Mon Sep 17 00:00:00 2001 From: Jun Siang Cheah Date: Sat, 25 Jul 2026 16:24:54 +0100 Subject: [PATCH] Refresh stale provider data and fix Vertex AI table leak Verified every provider against its live API/docs. Updates the hardcoded sections that nothing refreshes automatically: - Cerebras: free tier is now gpt-oss-120b, zai-glm-4.7 and gemma-4-31b at 5 RPM / 30K TPM / 1M TPH / 1M TPD (was 30 RPM / 60K TPM / 14,400 RPD, and still listed Llama 3.1 8B, which is gone). - OpenCode Zen: corrected two model names and added the four free models that were missing. - Google AI Studio: Gemini 3.5 Flash was reading gemini-3-flash's quota. Added Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, Gemma 4 31B and Gemma 4 26B A4B, which all have free-tier quota. - Mistral (La Plateforme): limits are set per-model and per-organization, not a flat 1 RPS / 500K TPM. Verified against the admin console. - Modal: the $5/month tier no longer exists; Starter is $30/month. - Alibaba Model Studio: note the 90-day validity and Singapore-only endpoint. - Scaleway: free tier also includes 60 minutes of audio transcription. - Vercel AI Gateway: note that the free tier covers a subset of models. Also fixes the dead Vertex AI block emitting an orphan into the README by guarding the closing tag with the same condition as the opening one. --- README.md | 38 ++++++++++++----------- src/pull_available_models.py | 58 +++++++++++++++++++++++++----------- 2 files changed, 61 insertions(+), 35 deletions(-) diff --git a/README.md b/README.md index 9047195..0af95ee 100644 --- a/README.md +++ b/README.md @@ -50,13 +50,10 @@ This lists various services that provide free access or credits towards API-base Models share a common quota. -- [Hermes 3 Llama 3.1 405B](https://openrouter.ai/nousresearch/hermes-3-llama-3.1-405b:free) -- [Llama 3.2 3B Instruct](https://openrouter.ai/meta-llama/llama-3.2-3b-instruct:free) -- [Llama 3.3 70B Instruct](https://openrouter.ai/meta-llama/llama-3.3-70b-instruct:free) -- [cognitivecomputations/dolphin-mistral-24b-venice-edition:free](https://openrouter.ai/cognitivecomputations/dolphin-mistral-24b-venice-edition:free) - [cohere/north-mini-code:free](https://openrouter.ai/cohere/north-mini-code:free) - [google/gemma-4-26b-a4b-it:free](https://openrouter.ai/google/gemma-4-26b-a4b-it:free) - [google/gemma-4-31b-it:free](https://openrouter.ai/google/gemma-4-31b-it:free) +- [inclusionai/ling-3.0-flash:free](https://openrouter.ai/inclusionai/ling-3.0-flash:free) - [nvidia/nemotron-3-nano-30b-a3b:free](https://openrouter.ai/nvidia/nemotron-3-nano-30b-a3b:free) - [nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free](https://openrouter.ai/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free) - [nvidia/nemotron-3-super-120b-a12b:free](https://openrouter.ai/nvidia/nemotron-3-super-120b-a12b:free) @@ -66,18 +63,18 @@ Models share a common quota. - [nvidia/nemotron-nano-9b-v2:free](https://openrouter.ai/nvidia/nemotron-nano-9b-v2:free) - [openai/gpt-oss-20b:free](https://openrouter.ai/openai/gpt-oss-20b:free) - [poolside/laguna-m.1:free](https://openrouter.ai/poolside/laguna-m.1:free) +- [poolside/laguna-s-2.1:free](https://openrouter.ai/poolside/laguna-s-2.1:free) - [poolside/laguna-xs-2.1:free](https://openrouter.ai/poolside/laguna-xs-2.1:free) -- [qwen/qwen3-coder:free](https://openrouter.ai/qwen/qwen3-coder:free) -- [qwen/qwen3-next-80b-a3b-instruct:free](https://openrouter.ai/qwen/qwen3-next-80b-a3b-instruct:free) -- [tencent/hy3:free](https://openrouter.ai/tencent/hy3:free) ### [Google AI Studio](https://aistudio.google.com) Data is used for training when used outside of the UK/CH/EEA/EU. + + @@ -85,6 +82,8 @@ Data is used for training when used outside of the UK/CH/EEA/EU. + + @@ -105,7 +104,7 @@ Models tend to be context window limited. * Free tier (Experiment plan) requires opting into data training * Requires phone number verification. -**Limits (per-model):** 1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month +**Limits:** Set per-model and per-organization — check [your limits page](https://admin.mistral.ai/plateforme/limits). As of July 2026 a new free account sees anywhere from 25,000 to 20,000,000 tokens/minute and 0.03 to 12.5 requests/second depending on the model. - [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/) @@ -131,6 +130,8 @@ HuggingFace Serverless Inference limited to models smaller than 10GB. Some popul Routes to various supported providers. +The free tier covers a subset of the model catalogue, with per-model rate limits. + **Limits:** [$5/month](https://vercel.com/docs/ai-gateway/pricing) @@ -140,15 +141,20 @@ AI gateway with curated models. Free models may use data for improvement. -- Big Pickle Stealth -- Nemotron 3 Super Free +- Big Pickle - DeepSeek V4 Flash Free +- MiMo-V2.5 Free +- Laguna S 2.1 Free +- Ling-3.0-flash Free +- North Mini Code Free +- Nemotron 3 Ultra Free ### [Cerebras](https://cloud.cerebras.ai/)
Model NameModel Limits
Gemini 3.6 Flash250,000 tokens/minute
20 requests/day
5 requests/minute
Gemini 3.5 Flash250,000 tokens/minute
20 requests/day
5 requests/minute
Gemini 3 Flash250,000 tokens/minute
20 requests/day
5 requests/minute
Gemini 3.5 Flash-Lite250,000 tokens/minute
500 requests/day
15 requests/minute
Gemini 3.1 Flash-Lite250,000 tokens/minute
500 requests/day
15 requests/minute
Gemini 2.5 Flash250,000 tokens/minute
20 requests/day
5 requests/minute
Gemini 2.5 Flash-Lite250,000 tokens/minute
20 requests/day
10 requests/minute
Gemini 2.5 Flash TTS10,000 tokens/minute
10 requests/day
3 requests/minute
Gemini Robotics-ER 1.6250,000 tokens/minute
20 requests/day
5 requests/minute
Gemini Robotics-ER 1.5250,000 tokens/minute
20 requests/day
10 requests/minute
Gemma 4 31B Instruct16,000 tokens/minute
14,400 requests/day
30 requests/minute
Gemma 4 26B A4B Instruct16,000 tokens/minute
14,400 requests/day
30 requests/minute
Gemma 3 27B Instruct15,000 tokens/minute
14,400 requests/day
30 requests/minute
Gemma 3 12B Instruct15,000 tokens/minute
14,400 requests/day
30 requests/minute
Gemma 3 4B Instruct15,000 tokens/minute
14,400 requests/day
30 requests/minute
- - + + +
Model NameModel Limits
gpt-oss-120b30 requests/minute
60,000 tokens/minute
900 requests/hour
1,000,000 tokens/hour
14,400 requests/day
1,000,000 tokens/day
Llama 3.1 8B30 requests/minute
60,000 tokens/minute
900 requests/hour
1,000,000 tokens/hour
14,400 requests/day
1,000,000 tokens/day
gpt-oss-120b5 requests/minute
30,000 tokens/minute
1,000,000 tokens/hour
1,000,000 tokens/day
zai-glm-4.75 requests/minute
30,000 tokens/minute
1,000,000 tokens/hour
1,000,000 tokens/day
gemma-4-31b5 requests/minute
30,000 tokens/minute
1,000,000 tokens/hour
1,000,000 tokens/day
### [Groq](https://console.groq.com) @@ -266,8 +272,6 @@ Extremely restrictive input/output token limits. - Qwen 2.5 Coder 32B Instruct - Qwen QwQ 32B - - ## Providers with trial credits @@ -318,13 +322,13 @@ Extremely restrictive input/output token limits. ### [Alibaba Cloud (International) Model Studio](https://bailian.console.alibabacloud.com/) -**Credits:** 1 million tokens/model +**Credits:** 1 million tokens/model, valid for 90 days (Singapore endpoint only) **Models:** [Various open and proprietary Qwen models](https://www.alibabacloud.com/en/product/modelstudio) ### [Modal](https://modal.com) -**Credits:** $5/month upon sign up, $30/month with payment method added +**Credits:** $30/month on the Starter plan **Models:** Any supported model - pay by compute time @@ -358,7 +362,7 @@ Extremely restrictive input/output token limits. ### [Scaleway Generative APIs](https://console.scaleway.com/generative-api/models) -**Credits:** 1,000,000 free tokens +**Credits:** 1,000,000 free tokens, plus 60 minutes of audio transcription **Models:** - BGE-Multilingual-Gemma2 diff --git a/src/pull_available_models.py b/src/pull_available_models.py index 531f669..09baeae 100644 --- a/src/pull_available_models.py +++ b/src/pull_available_models.py @@ -721,9 +721,14 @@ def main(): gemini_text_models = [ { - "id": "gemini-3-5-flash", + "id": "gemini-3.6-flash", + "name": "Gemini 3.6 Flash", + "limits": gemini_models.get("gemini-3.6-flash", {}), + }, + { + "id": "gemini-3.5-flash", "name": "Gemini 3.5 Flash", - "limits": gemini_models.get("gemini-3-flash", {}), + "limits": gemini_models.get("gemini-3.5-flash", {}), }, { "id": "gemini-3-flash-preview", @@ -731,7 +736,12 @@ def main(): "limits": gemini_models.get("gemini-3-flash", {}), }, { - "id": "gemini-3.1-flash-lite-preview", + "id": "gemini-3.5-flash-lite", + "name": "Gemini 3.5 Flash-Lite", + "limits": gemini_models.get("gemini-3.5-flash-lite", {}), + }, + { + "id": "gemini-3.1-flash-lite", "name": "Gemini 3.1 Flash-Lite", "limits": gemini_models.get("gemini-3.1-flash-lite", {}), }, @@ -765,6 +775,16 @@ def main(): "name": "Gemini Robotics-ER 1.5", "limits": gemini_models.get("gemini-robotics-er-1.5-preview", {}), }, + { + "id": "gemma-4-31b-it", + "name": "Gemma 4 31B Instruct", + "limits": gemini_models.get("gemma-4-31b", {}), + }, + { + "id": "gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B Instruct", + "limits": gemini_models.get("gemma-4-26b", {}), + }, { "id": "gemma-3-27b-it", "name": "Gemma 3 27B Instruct", @@ -814,7 +834,7 @@ def main(): "* Free tier (Experiment plan) requires opting into data training\n" ) model_list_markdown += "* Requires phone number verification.\n\n" - model_list_markdown += "**Limits (per-model):** 1 request/second, 500,000 tokens/minute, 1,000,000,000 tokens/month\n\n" + model_list_markdown += "**Limits:** Set per-model and per-organization — check [your limits page](https://admin.mistral.ai/plateforme/limits). As of July 2026 a new free account sees anywhere from 25,000 to 20,000,000 tokens/minute and 0.03 to 12.5 requests/second depending on the model.\n\n" model_list_markdown += "- [Open and Proprietary Mistral models](https://docs.mistral.ai/getting-started/models/models_overview/)\n" model_list_markdown += "\n" @@ -839,6 +859,7 @@ def main(): # --- Vercel AI Gateway --- model_list_markdown += "### [Vercel AI Gateway](https://vercel.com/docs/ai-gateway)\n\n" model_list_markdown += "Routes to various supported providers.\n\n" + model_list_markdown += "The free tier covers a subset of the model catalogue, with per-model rate limits.\n\n" model_list_markdown += "**Limits:** [$5/month](https://vercel.com/docs/ai-gateway/pricing)\n\n" model_list_markdown += "\n" @@ -846,23 +867,23 @@ def main(): model_list_markdown += "### [OpenCode Zen](https://opencode.ai/docs/zen/)\n\n" model_list_markdown += "AI gateway with curated models.\n\n" model_list_markdown += "Free models may use data for improvement.\n\n" - model_list_markdown += "- Big Pickle Stealth\n" - model_list_markdown += "- Nemotron 3 Super Free\n" + model_list_markdown += "- Big Pickle\n" model_list_markdown += "- DeepSeek V4 Flash Free\n" + model_list_markdown += "- MiMo-V2.5 Free\n" + model_list_markdown += "- Laguna S 2.1 Free\n" + model_list_markdown += "- Ling-3.0-flash Free\n" + model_list_markdown += "- North Mini Code Free\n" + model_list_markdown += "- Nemotron 3 Ultra Free\n" model_list_markdown += "\n" # --- Cerebras --- model_list_markdown += "### [Cerebras](https://cloud.cerebras.ai/)\n\n" model_list_markdown += "\n" + cerebras_free_limits = "5 requests/minute
30,000 tokens/minute
1,000,000 tokens/hour
1,000,000 tokens/day" cerebras_models = [ - { - "name": "gpt-oss-120b", - "limits_text": "30 requests/minute
60,000 tokens/minute
900 requests/hour
1,000,000 tokens/hour
14,400 requests/day
1,000,000 tokens/day" - }, - { - "name": "Llama 3.1 8B", - "limits_text": "30 requests/minute
60,000 tokens/minute
900 requests/hour
1,000,000 tokens/hour
14,400 requests/day
1,000,000 tokens/day" - }, + {"name": "gpt-oss-120b", "limits_text": cerebras_free_limits}, + {"name": "zai-glm-4.7", "limits_text": cerebras_free_limits}, + {"name": "gemma-4-31b", "limits_text": cerebras_free_limits}, ] for model in cerebras_models: model_list_markdown += ( @@ -947,7 +968,8 @@ def main(): limits_str = get_human_limits(model) model_list_markdown += f'\n' - model_list_markdown += "
Model NameModel Limits
{model['name']}{limits_str}
Free during preview
\n\n" + if vertex_llama_models or vertex_gemini_models or vertex_deepseek_models: + model_list_markdown += "\n\n" # --- Trial Providers Section Generation --- trial_list_markdown = "" @@ -1006,14 +1028,14 @@ def main(): { "name": "Alibaba Cloud (International) Model Studio", "url": "https://bailian.console.alibabacloud.com/", - "credits": "1 million tokens/model", + "credits": "1 million tokens/model, valid for 90 days (Singapore endpoint only)", "requirements": "", "models_desc": "[Various open and proprietary Qwen models](https://www.alibabacloud.com/en/product/modelstudio)", }, { "name": "Modal", "url": "https://modal.com", - "credits": "$5/month upon sign up, $30/month with payment method added", + "credits": "$30/month on the Starter plan", "requirements": "", "models_desc": "Any supported model - pay by compute time", }, @@ -1054,7 +1076,7 @@ def main(): # --- Scaleway Generative APIs (Trial - Table) --- if scaleway_models: trial_list_markdown += "### [Scaleway Generative APIs](https://console.scaleway.com/generative-api/models)\n\n" - trial_list_markdown += "**Credits:** 1,000,000 free tokens\n\n" + trial_list_markdown += "**Credits:** 1,000,000 free tokens, plus 60 minutes of audio transcription\n\n" trial_list_markdown += "**Models:**\n" for model in scaleway_models: trial_list_markdown += f"- {model['name']}\n"