Tinker: GLM 5.3 Fine-Tuning

Tinker, an AI model fine-tuning platform, has added GLM 5.3 (256K) to its training lineup, alongside other models such as Inkling, Nemotron-3.5-Lightning-30B-A3B, and Qwen3.8-27B, with prices per million tokens and an 80% discount on cached prefill tokens. The platform also offers serverless inference in beta for Inkling and Inkling-Small, with checkpoint storage charged at $0.10 per GB per month.

Models & Pricing All prices are per million tokens. Checkpoint storage is charged at $0.10 per GB per month. Training We provide an 80% discount on cached prefill tokens. | Model | Tinker ID | Context | Size | Arch | Type | PrefillCached: 80% discount | Sample | Train | |---|---|---|---|---|---|---|---|---| | Inkling 256K https://huggingface.co/thinkingmachines/Inkling Limited-time 50% discount Inkling-Small https://huggingface.co/thinkingmachines/Inkling-Small Limited-time 50% discount Inkling-Small 256K https://huggingface.co/thinkingmachines/Inkling-Small Limited-time 50% discount Nemotron-3.5-Lightning-30B-A3B https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 Limited-time 50% discount Nemotron-3.5-Lightning-30B-A3B 256K https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 Limited-time 50% discount Nemotron-3-Ultra-550B-A55B https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 Limited-time 50% discount Nemotron-3-Ultra-550B-A55B 256K https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 Limited-time 50% discount Nemotron-3-Super-120B-A12B https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 Limited-time 50% discount Nemotron-3-Super-120B-A12B 256K https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 Limited-time 50% discount Nemotron-3-Nano-30B-A3B https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 Limited-time 50% discount GLM-5.3 256K https://huggingface.co/zai-org/GLM-5.3 Kimi-K2.6 https://huggingface.co/moonshotai/Kimi-K2.6 Kimi-K2.6 128K https://huggingface.co/moonshotai/Kimi-K2.6 Qwen3.8-27B https://huggingface.co/Qwen/Qwen3.8-27B Qwen3.8-27B 256K https://huggingface.co/Qwen/Qwen3.8-27B Qwen3.6-35B-A3B https://huggingface.co/Qwen/Qwen3.6-35B-A3B Qwen3.6-27B https://huggingface.co/Qwen/Qwen3.6-27B Retiring September 2 ../model-deprecations/ Qwen3.5-397B-A17B https://huggingface.co/Qwen/Qwen3.5-397B-A17B Qwen3.5-397B-A17B 256K https://huggingface.co/Qwen/Qwen3.5-397B-A17B Qwen3.5-35B-A3B-Base https://huggingface.co/Qwen/Qwen3.5-35B-A3B-Base Qwen3.5-9B https://huggingface.co/Qwen/Qwen3.5-9B Qwen3.5-9B-Base https://huggingface.co/Qwen/Qwen3.5-9B-Base Qwen3.5-4B https://huggingface.co/Qwen/Qwen3.5-4B Qwen3-8B https://huggingface.co/Qwen/Qwen3-8B GPT-OSS-120B https://huggingface.co/openai/gpt-oss-120b GPT-OSS-120B 128K https://huggingface.co/openai/gpt-oss-120b GPT-OSS-20B https://huggingface.co/openai/gpt-oss-20b DeepSeek-V3.1 https://huggingface.co/deepseek-ai/DeepSeek-V3.1 Serverless Inference Beta Serverless inference is currently in beta and available for Inkling and Inkling-Small only. We do not recommend it for intensive production use until it is out of beta. If you're interested in production use, email us at email protected /cdn-cgi/l/email-protection 98ecf1f6f3fdead8ecf0f1f6f3f1f6fff5f9fbf0f1f6fdebb6f9f1a7ebedfaf2fdfbeca5cbfdeaeefdeaf4fdebebbdaaa8d1f6fefdeafdf6fbfdbdaaa8eff9f1ecf4f1ebecbef9f5e8a3faf7fce1a5cff0f1fbf0bdaaa8f7eafff9f6f1e2f9ecf1f7f6bdaaa8e1f7edbdaaa8eafde8eafdebfdf6ecbdabd9bda8d9bda8d9cff0f1fbf0bdaaa8f5f7fcfdf4ebbdaaa8e1f7edbdaaa8f6fdfdfcbdabd9bda8d9bda8d9dde0e8fdfbecfdfcbdaaa8eef7f4edf5fdbdabd9bda8d9bda8d9d4f9ecfdf6fbe1bdaaa8eafde9edf1eafdf5fdf6ecebbdabd9bda8d9bda8d9cdebfdbdaaa8fbf9ebfdbdabd9bda8d9bda8d9 to join the waitlist. Please include which models you need, your expected volume and latency requirements, and your use case. | Model | Tinker ID | Context | Prefill Input | Sample Output | |---|---|---|---|---| | Inkling https://huggingface.co/thinkingmachines/Inkling Pricing Terms Prefill : Processing input/prompt tokens forward pass only Cached prefill : The smaller price under each prefill price; applies to input tokens that hit the prompt cache 80% off Sample : Generating output tokens forward pass + sampling Train : Forward and backward pass for gradient computation Context : Maximum sequence length. Models with :peft: suffix support extended context at higher prices. Tinker ID : The exact string to pass to create lora training client base model=... or create sampling client base model=... MoE models are priced by active parameters, making them significantly more cost-effective than dense models of similar quality. Model Types Base : Raw pretrained models with no chat or instruction tuning. Best for post-training research or running the full post-training pipeline yourself. Reasoning : Always produce chain-of-thought before their answer. Highest intelligence, higher latency and token cost. Hybrid : Run in both thinking and non-thinking modes. They reason by default, but chain-of-thought can be disabled via a renderer or argument for faster, cheaper direct answers. Vision : Vision-language models that accept images alongside text. Shown as a + Vision suffix on the underlying type for example, Hybrid + Vision . Audio : Models that accept audio alongside text. Shown as a + Audio suffix on the underlying type. Architecture is either Dense all parameters active per token or MoE mixture-of-experts, only a subset of parameters active per token . MoE models are highlighted in amber. Choosing a Model Cost-effective : Use MoE models highlighted in amber Research/post-training : Use Base models Task-specific fine-tuning : Start with a Hybrid model Low latency : Use a Hybrid model with chain-of-thought disabled High intelligence : Use Reasoning or Hybrid models chain-of-thought Vision tasks : Use models with Vision in the type Machine-Readable Pricing If you want to use this data programmatically cost estimation, model pickers, dashboards , don't scrape the tables above. The data behind them is published as JSON alongside this page, and those files are the stable interface for scripts the tables' HTML is presentational and may change : models.json ../models.json : the Training table. One object per model with name , tinker id , context , size , arch , type , url , and per-million-token prices prefill , cached prefill , sample , train strings like "$0.374" . Temporarily discounted models also carry original price fields and a note . serverless.json ../serverless.json : the Serverless Inference table, with name , tinker id , context , url , input , cached input , output . For example: python import json, urllib.request url = "https://tinker-docs.thinkingmachines.ai/tinker/models.json" models = json.load urllib.request.urlopen url for m in models: print m "tinker id" , m "train" Retired Models These models have been retired and can no longer be used for training or inference, grouped by retirement date. See Model deprecations ../model-deprecations/ for the recommended replacement for each. July 12, 2026 Kimi: Kimi-K2.5 June 12, 2026 Qwen: Qwen3-235B-A22B-Instruct-2507 , Qwen3-VL-235B-A22B-Instruct , Qwen3.5-35B-A3B , Qwen3.5-27B , Qwen3-32B , Qwen3-30B-A3B , Qwen3-30B-A3B-Instruct-2507 , Qwen3-VL-30B-A3B-Instruct , Qwen3-30B-A3B-Base , Qwen3-8B-Base , Qwen3-4B-Instruct-2507 Llama: Llama-3.3-70B-Instruct , Llama-3.1-70B , Llama-3.1-8B , Llama-3.1-8B-Instruct , Llama-3.2-3B , Llama-3.2-1B DeepSeek: DeepSeek-V3.1-Base Kimi: Kimi-K2-Thinking