Model catalog

Everything we serve, with the numbers most providers leave out. Quantization is a spec, not a footnote — every model runs with TextCLF Quant (TQ), and you can see the exact precision, latency, and price before you spend a cent.

4 models serving now

Llama 3.3 70B Instruct

Live

meta-llama/Llama-3.3-70B-Instruct

Parameters

70B

Max context

128K

Meta’s flagship dense model, tuned for instruction following, tool use, and multilingual chat. TQ 4-bit keeps it within a fraction of a point of full precision on reasoning benchmarks while cutting the memory it takes to serve.

ProviderTextCLF

Context128K

P50 TTFT544ms

Throughput6.6 tok/s

Uptime99.99%

QuantizationTQ 4-bit

Input /M$0.08

Output /M$0.30

Cached input /M—

generaltool-usemultilingual

meta-llama/Llama-3.3-70B-Instruct

Llama 3.1 8B Instruct

Live

meta-llama/Llama-3.1-8B-Instruct

Parameters

Max context

128K

The fastest, cheapest model in the catalog. Ideal for classification, extraction, and high-volume routing where latency and price matter more than frontier reasoning.

ProviderTextCLF

Context128K

P50 TTFT140ms

Throughput21.2 tok/s

Uptime99.99%

QuantizationTQ 4-bit

Input /M$0.018

Output /M$0.038

Cached input /M—

fastcheapclassification

meta-llama/Llama-3.1-8B-Instruct

Qwen3.8 27B

Live

Qwen/Qwen3.8-27B

Parameters

27B

Max context

262K

A balanced mid-tier model with strong coding and math performance and a long context window. The right default when 8B is too small and 70B is more than the task needs.

ProviderTextCLF

Context262K

P50 TTFT393ms

Throughput21.3 tok/s

Uptime99.98%

QuantizationTQ 4-bit

Input /M$0.30

Output /M$2.50

Cached input /M$0.03

balancedcodinglong-context

Qwen/Qwen3.8-27B

Qwen3 Coder Next

Live

Qwen/Qwen3-Coder-Next

Parameters

—

Max context

262K

A coding-focused Qwen model optimized for software engineering, code generation, and agentic development workflows. Served with TextCLF Quant compression.

ProviderTextCLF

Context262K

P50 TTFT—

Throughput—

Uptime—

QuantizationTQ 4-bit

Input /M$0.10

Output /M$0.78

Cached input /M$0.03

codingagentslong-context

Qwen/Qwen3-Coder-Next

DeepSeek V4 Flash

Coming soon

deepseek-ai/DeepSeek-V4-Flash-0731

Parameters

MoE (sparse)

Max context

A fast mixture-of-experts model tuned for reasoning and coding, activating only a fraction of its parameters per token. TQ 4-bit makes frontier-class reasoning practical to serve at open-weight prices. Coming soon — join the waitlist from your dashboard.

ProviderTextCLF

Context1M

P50 TTFT—

Throughput—

Uptime—

QuantizationTQ 4-bit

Input /M$0.05

Output /M$0.12

Cached input /M$0.01

moereasoningcoding

deepseek-ai/DeepSeek-V4-Flash-0731

MiMo V2.5

Coming soon

XiaomiMiMo/MiMo-V2.5

Parameters

—

Max context

—

Xiaomi MiMo V2.5 served with TextCLF Quant compression. Coming soon.

ProviderTextCLF

Context—

P50 TTFT—

Throughput—

Uptime—

QuantizationTQ 4-bit

Input /M$0.11

Output /M$0.23

Cached input /M$0.00255

reasoningcoming-soon

XiaomiMiMo/MiMo-V2.5

Prices per 1M tokens (USD). 100+ additional models available on request.

Run the proof yourself

Don't take our word for it.

Spin up a TQ 4-bit model on your own GPU with Docker and vLLM. Try the same weights locally, test your real workloads, and decide on the evidence — before you commit to a hosted endpoint.

Browse TextCLF on Hugging Face →

Qwen3 Coder Next

textclf/Qwen3-Coder-Next-TQ-4bit

sudo docker run --rm --gpus all -p 8000:8000 -v ~/.cache/huggingface:/root/.cache/huggingface docker.io/textclf/tq-quant:4bit-v1 vllm serve textclf/Qwen3-Coder-Next-TQ-4bit --max-num-batched-tokens 8192 --enable-auto-tool-choice --tool-call-parser qwen3_coder --quantization tq_quant --dtype float16 --trust-remote-code --generation-config vllm --gpu-memory-utilization 0.90 --enable-prefix-caching --enable-prompt-tokens-details --max-num-seqs 16 --max-cudagraph-capture-size 16

Qwen3.8 27B

textclf/Qwen3.8-27B-TQ-4bit

sudo docker run --rm --gpus all -p 8000:8000 -v ~/.cache/huggingface:/root/.cache/huggingface docker.io/textclf/tq-quant:4bit-v1 vllm serve textclf/Qwen3.8-27B-TQ-4bit --quantization tq_quant --dtype float16 --trust-remote-code --kv-cache-dtype turboquant_4bit_nc --generation-config vllm --gpu-memory-utilization 0.90 --enable-auto-tool-choice --tool-call-parser qwen3_coder --reasoning-parser qwen3 --mm-encoder-tp-mode data --max-num-batched-tokens 8192 --enable-prefix-caching --enable-prompt-tokens-details

Llama 3.3 70B Instruct

textclf/Llama-3.3-70B-Instruct-TQ-4bit

sudo docker run --rm --gpus all -p 8000:8000 -v ~/.cache/huggingface:/root/.cache/huggingface docker.io/textclf/tq-quant:4bit-v1 vllm serve textclf/Llama-3.3-70B-Instruct-TQ-4bit --quantization tq_quant --generation-config vllm --dtype float16 --trust-remote-code --kv_cache_dtype turboquant_4bit_nc --gpu_memory_utilization 0.90 --tool-call-parser llama3_json --enable-auto-tool-choice --max-num-batched-tokens 8192 --max-num-seqs 256 --enable-prefix-caching --enable-prompt-tokens-details

Llama 3.1 8B Instruct

textclf/Llama-3.1-8B-Instruct-TQ-4bit

sudo docker run --rm --gpus all -p 8000:8000 -v ~/.cache/huggingface:/root/.cache/huggingface docker.io/textclf/tq-quant:4bit-v1 vllm serve textclf/Llama-3.1-8B-Instruct-TQ-4bit --quantization tq_quant --dtype float16 --generation-config vllm --trust-remote-code --kv_cache_dtype turboquant_4bit_nc --gpu_memory_utilization 0.90 --tool-call-parser llama3_json --enable-auto-tool-choice --max-num-batched-tokens 8192 --max-num-seqs 256 --enable-prefix-caching --enable-prompt-tokens-details --speculative-config {"model":"RedHatAI/Llama-3.1-8B-Instruct-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}

These are the exact quantized weights behind our hosted inference. Run them on your own GPU, point your existing OpenAI client at localhost, and make the comparison with your own prompts.

Requires Docker, a CUDA-capable GPU, and a local Hugging Face cache. Each command serves an OpenAI-compatible endpoint on port 8000.