Inkling
Thinking MachinesReleased Jul 15, 2026Open weights · Apache 2.0inkling
- Intelligence (Epoch)
- 149 · 44th of 109
- Reasoning
- yes, with effort levels
- Context
- 1M tokens
- Input
- text, image, audio
- API features
- tools
Read Nov 3, 11:57 UTC from the hosts' published offers and OpenRouter, spec sheet from Thinking Machines
Free right now
- Free OpenRouter, served by Thinking Machines · until Nov 4 · 20 req/min, 50 req/day, 1,000 req/day with $10 of credits
Checked Nov 3 · every free model on Deals
The cheapest host, DeepInfra, serves fp8 weights (maker: bf16) and serves a 512K context (maker: 1M). Weights and context per host
| Provider | Input | Output | Cache ↑ | tok/s | First token | Uptime 24h | Checked |
|---|---|---|---|---|---|---|---|
| DeepInfrafp8 (bf16) · 512K ctx (1M) | $0.95 | $4.05 | $0.16 | — | — | — | Oct 22 |
| BaseTenprecision not stated · 1M ctx | $1.00 | $4.05 | $0.17 | — | — | — | Sep 25 |
| Together AInvfp4 (bf16) · 512K ctx (1M) | $1.00 | $4.05 | $0.17 | 87 | 0.46 s | 66.3% | Nov 3 |
3 of 3 providers · prices per million tokens, read Nov 3, 11:57 UTC · speed: OpenRouter, median of the last 30 min, read Nov 3, 11:57 UTC · uptime: OpenRouter, last 24 h · under each host: weights precision, context and max output as served, the maker's value in brackets when the host's differs
| Provider | Weights | Context | Max output | Tools | Structured outputs | Reasoning |
|---|---|---|---|---|---|---|
| DeepInfra | fp8 | 512K | not stated by the host | — | — | — |
| BaseTen | not stated by the host | 1M | not stated by the host | — | — | — |
| Together AI | nvfp4 | 512K | not stated by the host | yes | no | yes |
Weights, context, max output: from the host where its terms allow, else as declared to OpenRouter (hover for source and date) · features as declared to OpenRouter, not verified by us
Inkling spec sheet, field by field, with sources
Inkling spec sheetmaker-declared
What Thinking Machines declares, from its own weights and docs. What each host serves is in the offers: both can be true.
- Context window
- not published by the maker
- Parameters
- 952,377,623,626≈ 952.4B
- Active experts
- 6 of 256 active per token
- Weight precision
- bfloat16
- License
- Apache 2.0 — open weights, permissivecurated · read 2026-09-16Scope: the license of the published weights only — not the maker's acceptable-use policy, nor the terms of the host you call.
- Knowledge cutoff
- to verify
- Modalities
- text · image · audio
- Max output (official)
- not published by the makermaker's docs · read 2026-10-23
- Release date
- to verify≈ July 15, 2026 third-party catalogue — not the maker's figure
- Maker lifecycle status
- to verify
Also declared in the weights
- Architecture class
- InklingForConditionalGeneration
- Layers
- 66
- Vocabulary
- 201,024 tokens
- Weights first published
- July 14, 2026
- Weights last updated
- July 23, 2026
thinkingmachines/Inkling · config.json @ 828496e · read 2026-10-08 · 5 of 10 fields published by the maker
| Provider | Chatbot 1,000 in / 500 out | RAG / long context 8,000 in / 500 out | Batch summarize 4,000 in / 1,000 out |
|---|---|---|---|
| DeepInfra | $2.98 | $9.62 | $7.85 |
| BaseTen | $3.02 | $10.02 | $8.05 |
| Together AI | $3.02 | $10.02 | $8.05 |
What each host bills for 1,000 requests of a representative job
Your own workload
Intelligence for the price
20 models score higher for the same price or less; it ranks 21 of the 78 models at its price or below.
Epoch Capabilities Index against the lowest input + output price per M, log scale · prices Nov 3. Score: Epoch AI, CC BY 4.0, 149 with an interval of 146 to 150.
Against its rivals
The best of its price range and the top score
- Inklingthis model$5.00149
- GPT-6 Astratop score$60.00169
- Gemini 3.8 Flashsame price range$4.50158
- Gemini 3.7 Flashsame price range$4.50158
Same level for 31× less: Qwen3.7 Flash at Alibaba, $0.16 per M input + output · their Epoch intervals overlap (144 against 149).
Bars: Epoch Capabilities Index, drawn from 100 · price: lowest input + output per M.
Run it yourself
open weights, Apache 2.0 · too large for the 14 workstations in our guideSelf-host guide ›- Parameters
- 975B
- 41B active per token (MoE)
- Memory needed
- 585 GB
- in INT4 · 1170 GB in INT8
- Smallest cloud setup
- 8× H100
- 640 GB · $21.12/h, about $15,418 a month
- Worth it above
- 3B tokens
- a month, against the cheapest API at $5.00 per M
| Quantization | Memory | Smallest cloud GPU | Quality |
|---|---|---|---|
| INT4 (Q4) | 585 GB | 8× H100 · $21.12/h | smallest footprint, minor quality loss |
| INT8 (Q8) | 1170 GB | multi-node, over 640 GB | near-lossless |
| FP16 (full) | 2340 GB | multi-node, over 640 GB | reference quality |
At home: 585 GB in INT4 is beyond every machine we price, so it means a multi-GPU server.
Memory ≈ parameters × bytes × 1.2 · cloud: on-demand, 24/7 · builds: GGUF, AWQ, MLX, GPTQ, Ollama · break-even calculator
Other Thinking Machines models · all its models, hosts and history ›
- Inkling Small2 providers · from $1.65