Head to head · Open-weight models you run yourself · October 2026 research run

Inkling 2 Small vs Recursive Flint

Recursive Flint scores 61 (C) on agent readiness against Inkling 2 Small's 58.8 (C), and leads in 2 of 4 scored categories. Inkling 2 Small leads on risk. Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 6472 open weights comparisons

Which one, for what

Inkling 2 Small C

Best for Teams with 512 GB or more of datacentre GPU memory who want Thinking Machines' newest open multimodal agent model and can run a runtime's pre-release image.

Ahead on

  • Risk, 65 against 60

Watch for

The smaller checkpoint needs about 472 GB, and Thinking Machines says 500 GB of GPU memory

Recursive Flint C

Best for Agents doing repository work on a 32 to 48 GB Mac or NVIDIA box that want a sparse coder with documented tool calls and a plain licence.

Ahead on

  • Reach, 85 against 40
  • Agent interface, 90 against 85

Watch for

No independent benchmark results yet, every published score is Recursive's

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Reach Recursive Flint +45

Inkling 2 Small 40

Memory 0 of 30, the smallest build is Thinking Machines' NVFP4 checkpoint at 393.5 GB, about 472 GB with a fifth added, above the 256 GB line. Runtimes 0, no released version of the six runs it; vLLM and SGLang serve it from their own images and Transformers has it on its main branch only. Builds 15, two precisions from the publisher (NVFP4 and MXFP8). Terms 10, Apache 2.0 with a binding acceptable use policy on top, read as a custom licence. Download 10, no gate. Hosted 5, Thinking Machines' own Tinker serves it as beta serverless inference, and Together is live on Hugging Face. 0 + 0 + 15 + 10 + 10 + 5 = 40.

Recursive Flint 85

Memory 20 of 30, Recursive's MLX INT4 build is 22.6 GB, or 27.1 GB with a fifth added, which fits 32 GB, and every other 4-bit build lands in the same band. Runtimes 25 of 25, released llama.cpp (b11803), Ollama 0.40.8, LM Studio's catalogue, Transformers v5.21 and vLLM v0.33.0 run it, while SGLang has only a development image. Builds 15, MLX INT4 plus Q4_1 and NVFP4 GGUFs from Recursive. Terms 15, Apache 2.0 with nothing added. Download 10, nothing is gated. Hosted 0, OpenRouter lists a free endpoint with Recursive as its provider, but it wasn't serving on 3 November 2026, with 0 per cent uptime over the last day.

Risk Inkling 2 Small +5

Inkling 2 Small 65

File safety 25, safetensors only and no repository code to run. Provenance 15, the files are in the organisation Thinking Machines links to, with a SHA-256 on every LFS file. Safety evaluation 10, Thinking Machines' own StrongREJECT, FORTRESS, HoneyBench and DistressBench results and described outside red-teaming, but nothing an outside party published or a harness to rerun. Agent guidance 5, the card names safeguards to add but doesn't say what its refusals cover, with nothing on prompt injection. Licence risk 5, 10 off for a use policy that reaches beyond the law and that can change at any time. Security programme 5, a security.txt and a disclosure policy, with public advisories unchecked and scored as absent. 25 + 15 + 10 + 5 + 5 + 5 = 65.

Recursive Flint 60

File safety 15 of 25, safetensors and GGUF only, but the card's documented BF16 route installs the wheel shipped in the repository. Provenance 15, the files are in the RecursiveSI organisation that Recursive's own engine downloads from, with a checksum on every file. Safety evaluation 0, none on the cards or in the announcement post. Agent guidance 10, the card and serving README say to enforce permissions around tool use and to strip tool tags from untrusted tool output, but there's no trained instruction hierarchy. Licence risk 15, plain Apache 2.0. Security programme 5, a reporting address on the card and a private reporting route for the engine, with no public advisories yet.

Agent interface Recursive Flint +5

Inkling 2 Small 85

Chat template 20, a Jinja file in the repository. Tool-calling format 25, Thinking Machines' tml-renderers guide shows how tools are declared and how calls come back for this model, and its package ships the parser. Trained tool calling 20, several calls in one turn. Structured output 0, not documented. Reasoning control 10, thinking effort is set per request from none to xhigh, and the reasoning comes back in reasoning_content. Context 10, up to 1M tokens. 20 + 25 + 20 + 0 + 10 + 10 = 85.

Recursive Flint 90

Chat template 20, a Jinja file in the repository and inside both GGUFs. Tool-calling format 25, Recursive's serving README documents how tools go in, how calls come back and the flint parser. Trained tool calling 20, several calls a turn. Structured output 15, response_format with JSON Schema is documented. Reasoning control 0, the model decides when to reason and Recursive says there's no switch or budget. Context 10, 262,144 tokens with no position scaling.

Score by category

CategoryWeight this runInkling 2 SmallRecursive FlintEdge
Capability35%pendingpendingpendingnot scored in this run
Reach25%254085Recursive Flint +45
Risk20%206560Inkling 2 Small +5
Agent interface10%108590Recursive Flint +5
Documentation and upkeep10%106767even
Negative events≤1500
TotalC 58.8/100C 61/100

Facts side by side

FactInkling 2 SmallRecursive Flint
KindOpen-weight modelOpen-weight model
VendorThinking Machines LabRecursive
Hosted endpointno (local only)no (local only)
Transports
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0Apache-2.0
Read-only variant documentednono
llms.txtnono
Last release2026-11-122026-10-21
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Agent reviewsnone3.5/5 (4)

How the scores have moved

Verdicts

Inkling 2 Small

Thinking Machines' newest open model, a 594B mixture-of-experts generalist with images in, several tool calls a turn and per-request thinking effort. The NVFP4 checkpoint needs about 472 GB, and no released runtime runs it yet, so today that means vLLM's or SGLang's own images, or Tinker and Together hosted. The use policy binds on download and can change.

Recursive Flint

A small-active coding cousin for agents with 32 to 48 GB, built for codebase work with parallel tool calls and JSON output. Released llama.cpp, Ollama, vLLM and Transformers run it, while Recursive's free OpenRouter endpoint wasn't serving on 3 November 2026. Every capability number so far is Recursive's own, there are no safety results, and reasoning can't be turned off.

Before you call either

Inkling 2 Small

  1. No released runtime runs it yet. Serve the NVFP4 checkpoint from SGLang's dev-inkling-2-small image with the cookbook's recipe for your hardware, or from vLLM's vllm/vllm-openai:inkling-2-small image
  2. In vLLM pass --reasoning-parser inkling --enable-auto-tool-choice --tool-call-parser inkling. In SGLang the tool-call parser is inkling2 and the reasoning parser inkling
  3. Transformers' main branch loads only thinkingmachines/Inkling-2-Small-MXFP8, the 622.8 GB checkpoint, and no release has the code yet
  4. Pass reasoning_effort as none, minimal, low, medium, high (the default), xhigh or max, or a number from 0 to 0.99
  5. Effort none conditions the model toward no reasoning but doesn't forbid it, and higher effort needs a larger max_tokens
  6. Tool call arguments must reach the template as an object, not a JSON string, or it raises an error
  7. Images should be 40 to 4096 px a side. The model takes no audio, and the template raises an error for an audio part
  8. Use guided decoding or validate JSON yourself, since structured output isn't documented
  9. Without the hardware, call it on Tinker's OpenAI-compatible endpoint with a Tinker key, at $0.30 input and $1.00 output per million tokens as read on 6 November 2026
  10. Read the Model Acceptable Use Policy before deploying, it binds on download and bars some uses that are legal

Recursive Flint

  1. Quickest start is ollama run flint (Ollama 0.40.8 or later), about 23 GB of files; Ollama's Flint support is in its MLX runner, so on Linux prefer llama.cpp
  2. With llama.cpp use release b11803 or later: llama-server -hf RecursiveSI/flint-int4-gguf --alias flint -ngl all -c 131072 -np 1 -fa on --jinja --temp 1 --top-p 0.95 --top-k 20 --min-p 0, and keep --jinja on
  3. Set max_tokens to at least 32768 for hard tasks, reasoning counts against it and can't be switched off, and finish_reason length can mean no answer or tool call came out
  4. Send reasoning_content back on assistant messages when replaying the conversation, as Recursive's cards ask
  5. Strip <tool_call>, <tool_response>, <arg_key> and <arg_value> from untrusted tool output, the template doesn't escape them and a copied </tool_call> can add a call
  6. Message content must be a string, flatten content-part arrays first
  7. Recursive's SGLang servers listen on 0.0.0.0 with no key, publish the port on localhost only or pass --api-key
  8. Check OpenRouter's endpoints list before calling recursive/flint-v1.0:free, since its only endpoint (Recursive, BF16) showed 0 per cent uptime over the last day on 3 November 2026
  9. vLLM needs v0.33.0 or later and Transformers v5.21 or later; released SGLang has no Flint model

Questions

Which is better for AI agents, Inkling 2 Small or Recursive Flint?

Recursive Flint scores 61 (C) on agent readiness against Inkling 2 Small's 58.8 (C), and leads in 2 of 4 scored categories. Inkling 2 Small leads on risk.

Other comparisons with Inkling 2 Small or Recursive Flint

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.