Jais 3 vs SWE-2 Flash

Jais 3 65

Model reading of the checklist (tool use, structured output, caching, context, throughput, defaults and SDKs, errors). OpenAI-format tools come back as tool_calls in three selectable wire formats, with a multi-turn loop example. Forced tool_choice and parallel calls aren't documented (15 of 20). JSON mode and schemas only through each runtime's guided decoding, not documented by the makers (5 of 15). Prefix caching is the runtime's, and SGLang's page covers HiCache (7 of 10). 524,288 tokens on five of six models, 131,072 on the 0.9B (14 of 15). Uno adapters claim 2.15 times faster decoding for 0.9B to 32B but need IFM's own engine. SGLang lists n-gram speculative decoding and PD disaggregation. No batch mode of its own (7 of 10). Any OpenAI SDK works against a local server, and the 0.9B to 7B fit one GPU. But the cards ask for high effort and 32,768 output tokens on every request, the template inserts an identity system prompt when none is sent, and earlier assistant turns must carry a string thinking field (8 of 15). The template raises named errors for bad options, string-encoded tool arguments and malformed schemas. Server errors are the runtime's (9 of 15). 15 + 5 + 7 + 14 + 7 + 8 + 9 = 65.

SWE-2 Flash 10

Model reading (tool use, structured output, caching, context, batch, SDKs). Tool use is Devin's harness, not something a caller configures; no context length, output limit or cache price is published for SWE-2 Flash (5 of 25). No documented output-size or reasoning controls for this model; SWE-2's effort levels aren't stated for Flash (0 of 20). No model-level error documentation (0 of 20). No idempotency or retry guidance (0 of 20). One flag or picker selects it and devin -p runs a single turn and prints to stdout, but there is no SDK or direct endpoint (5 of 15). 5 + 0 + 0 + 0 + 5 = 10.

Jais 3 60

Self-hosted rule. Apache-2.0 weights with nothing to buy and no hosted option from the makers (20). Free to run with no card (20). Downloads are ungated and anonymous, so an agent can fetch the weights with no sign-up (20). No payment protocol (0). 20 + 20 + 20 + 0 = 60.

SWE-2 Flash 20

Published rubric. No payment protocol (0). Plan prices are public, but no per-token price for SWE-2 Flash is on a readable page; the docs' cost table loads in the browser and the post gives only 'one-fifth the price' of SWE-2 (10). Free through 30 November 2026 on Devin Desktop and CLI, but the docs say Free tier users only have SWE-1.6, so whether the Free plan reaches it, and without a card, isn't confirmed; the in-between value used on the devin listing (10). A person signs in through a browser (0). 0 + 10 + 10 + 0 = 20.

Jais 3 54

Read for a model the owner serves. No specification of its own. The contract is the Jinja chat template, which validates tool JSON Schemas (required fields must exist in properties) and rejects unknown options, plus the OpenAI-compatible servers in vLLM and SGLang with the k2_horizon parsers (10 of 25). The cards are Markdown on Hugging Face, and SGLang's docs have an llms.txt, but no llms.txt or Markdown docs from IFM were found and ifm.ai couldn't be read (5 of 10). Each card states intended use, four out-of-scope uses, limitations and per-size serving advice (15 of 20). reasoning_effort and tool_call_format are enums enforced with errors, and tools take JSON Schema. The 375B card describes low as no reasoning and omits medium, which the template accepts, and structured output isn't documented (10 of 15). Python examples for a reasoning call and a tool loop, commands for five runtimes, and named template errors, but no error reference for a served model (11 of 15). No changelog or version scheme. Training-stage branches and public commit history (3 of 15). 10 + 5 + 15 + 10 + 11 + 3 = 54.

SWE-2 Flash 29

Model reading. No model API, so no OpenAPI entry; the Devin v3 OpenAPI file lists SWE-2 agents only, and no model card was found (0). docs.devin.ai serves llms.txt and Markdown copies of every page (10). The launch post says what it is for (automations, code review, feature development, documentation) and the CLI docs say when to pick a SWE model, but the models pages don't list SWE-2 Flash yet (6 of 20). No documented model ID, reasoning settings, context or output limits to type (0 of 15). The CLI pages show --model and /model examples, with no model-specific errors (3 of 15). Named, dated model versions on Cognition's blog and public release notes, but no release note for SWE-2 Flash yet; the newest is 2 November (10 of 15). 0 + 10 + 6 + 0 + 3 + 10 = 29.

Jais 3 31

Read as software the owner runs. No account or key, ungated downloads, and server authentication is the owner's to add (12 of 30). A model that calls whatever tools the operator wires in. The cards rule high-stakes decisions without human oversight out of scope, but nothing in the model or template enforces approval (6 of 20). The template renders tool results in their own tool role, but string results go in as raw text (only non-string content is serialised with tojson), no escaping of special-token sequences in caller or tool text was found, and no injection guidance or results are published (4 of 15). No call log of its own. Visibility depends on the serving stack (5 of 15). Every launch command passes --trust-remote-code, so modeling_k2_horizon.py from the repository runs on the host. We read it and found only torch and transformers imports. No security.txt on ifm.ai (404 on 17 October, not re-read), no security contact on the cards and no safety evaluation found. Since 25 October the cards link a technical report on ifm.ai, which answered a Cloudflare bot check on 4 November, so what it covers is unread (4 of 20). 12 + 6 + 4 + 5 + 4 = 31.

SWE-2 Flash 53

Read as a hosted model reached through Devin's local clients. The CLI signs in through a browser, or the enterprise identity provider, and stores an API token in credentials.toml that by default is persistent and does not expire and carries the user's access, with no scopes found for it (10 of 30). Devin CLI documents an OS-level sandbox with network filtering, hooks that can block tool calls, and enterprise model allowlists and team settings (14 of 20). Guardrails against prompt injection are Enterprise only and cover user messages, as on the devin listing (8 of 15). Enterprise analytics count tool calls and consumption by model and user; no per-user audit log for the CLI was found (8 of 15). SOC 2 Type II and ISO/IEC 27001:2022, a disclosure address, no bug bounty and no security.txt, per the devin listing (13 of 20). Training: self-serve customer data may train Cognition's models by default, with an opt-out on paid plans that also turns on zero data retention at model providers. Cognition withheld the weights over cyber misuse, which is recorded as a fact, not scored. 10 + 14 + 8 + 8 + 13 = 53.

Jais 3 62

Read for an open-weight model released on 15 October 2026. The last change to the template or code was on 15 October, within 30 days. The cards gained a technical report link on 25 October and a citation update on 27 October, which change no file a runtime loads (30). One family release, with dated builds before it (FP8 from 29 September, MLX from 1 October, GGUF on 12 and 13 October) and template fixes on 14 and 15 October, but no deprecation notices or successor history to judge. K2 Horizon came on 1 September and Jais 2 in December 2025 (10 of 20, the model reading of this line, held). One outside question, on MoVA-36B-A4B on 26 October, had no reply on 4 November, and IFM's GitHub repositories weren't read. The template fixes match the problem IFM's MLX cards describe (7 of 25, held). Recipes in vLLM's and SGLang's own docs, K2 Horizon support in llama.cpp's tagged builds from b11454 and FP8, GGUF and MLX builds from IFM, but no native mlx-lm support (11 of 15). Current dependencies (Transformers 5.13 to 5.15, vLLM 0.30), with no CI visible (4 of 10). 30 + 10 + 7 + 11 + 4 = 62.

SWE-2 Flash 53

Model reading. Released 4 November 2026 (30). Churn and notice in place of release counts: SWE-1.6, SWE-1.7, SWE-1.7 Lightning, SWE-2 (10 September) and now SWE-2 Flash in a few months, with no model deprecation notice period found and short names that move to the newest model (10 of 20). Public release notes and a support address, no public tracker (10 of 25). No official SDK for the model (0 of 15). The docs are kept current but don't yet cover this model (3 of 10). 30 + 10 + 10 + 0 + 3 = 53.

Jais 3 55

Apache-2.0 in every card's metadata, with staged training checkpoints as branches and training logs linked on Weights & Biases (not opened). There is no LICENSE file and the code repository is still 'coming soon'. A technical report is linked from 25 October but sits behind a bot check, so its data description is unread (24 of 30). Inputs stay on the owner's hardware because nothing is hosted, but no statement says so and no privacy document governs the weights (15 of 30). No deprecation policy, dated notices or version scheme (0 of 20). Safetensors weights carry no code, and the remote modelling and configuration code we read imports only torch, transformers and huggingface_hub, with no network calls. No telemetry statement (12 of 20). 24 + 15 + 0 + 12 = 51.

SWE-2 Flash 64

Editorial half. Closed model under Cognition's platform terms; the weights aren't public and the sharing programme isn't described yet (15). Terms, privacy policy, security page and DPA agree, with training on self-serve data by default and no retention periods, as read for the devin listing (18). The API versioning policy covers API endpoints, not models, and no model deprecation notice was found (5 of 20). Sub-processors listed with purpose and country, per the devin listing's 8 October check (18). 15 + 18 + 5 + 18 = 56.